{"input": "Recognition of proper nouns in Japanese text has been studied as a part of the more general problem of morphological analysis in Japanese text processing -LRB- -LSB- 1 -RSB- -LSB- 2 -RSB- -RRB- .", "output": {"relations": {"part of": [{"head": {"text": "Recognition of proper nouns", "start": 0, "end": 27}, "tail": {"text": "morphological analysis", "start": 103, "end": 125}}, {"head": {"text": "proper nouns", "start": 15, "end": 27}, "tail": {"text": "Japanese text", "start": 31, "end": 44}}], "used for": [{"head": {"text": "morphological analysis", "start": 103, "end": 125}, "tail": {"text": "Japanese text processing", "start": 129, "end": 153}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "It has also been studied in the framework of Japanese information extraction -LRB- -LSB- 3 -RSB- -RRB- in recent years .", "output": {"relations": {"used for": [{"head": {"text": "Japanese information extraction", "start": 45, "end": 76}, "tail": {"text": "It", "start": 0, "end": 2}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our approach to the Multi-lingual Evaluation Task -LRB- MET -RRB- for Japanese text is to consider the given task as a morphological analysis problem in Japanese .", "output": {"relations": {"used for": [{"head": {"text": "approach", "start": 4, "end": 12}, "tail": {"text": "Multi-lingual Evaluation Task -LRB- MET -RRB-", "start": 20, "end": 65}}, {"head": {"text": "Multi-lingual Evaluation Task -LRB- MET -RRB-", "start": 20, "end": 65}, "tail": {"text": "Japanese text", "start": 70, "end": 83}}, {"head": {"text": "morphological analysis problem", "start": 119, "end": 149}, "tail": {"text": "task", "start": 109, "end": 113}}, {"head": {"text": "Japanese", "start": 70, "end": 78}, "tail": {"text": "morphological analysis problem", "start": 119, "end": 149}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our morphological analyzer has done all the necessary work for the recognition and classification of proper names , numerical and temporal expressions , i.e. Named Entity -LRB- NE -RRB- items in the Japanese text .", "output": {"relations": {"used for": [{"head": {"text": "morphological analyzer", "start": 4, "end": 26}, "tail": {"text": "recognition and classification of proper names , numerical and temporal expressions", "start": 67, "end": 150}}], "hyponym of": [{"head": {"text": "Named Entity -LRB- NE -RRB- items", "start": 158, "end": 191}, "tail": {"text": "proper names , numerical and temporal expressions", "start": 101, "end": 150}}], "part of": [{"head": {"text": "Named Entity -LRB- NE -RRB- items", "start": 158, "end": 191}, "tail": {"text": "Japanese text", "start": 199, "end": 212}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Amorph recognizes NE items in two stages : dictionary lookup and rule application .", "output": {"relations": {"used for": [{"head": {"text": "Amorph", "start": 0, "end": 6}, "tail": {"text": "NE items", "start": 18, "end": 26}}], "part of": [{"head": {"text": "dictionary lookup", "start": 43, "end": 60}, "tail": {"text": "Amorph", "start": 0, "end": 6}}, {"head": {"text": "rule application", "start": 65, "end": 81}, "tail": {"text": "Amorph", "start": 0, "end": 6}}], "conjunction": [{"head": {"text": "dictionary lookup", "start": 43, "end": 60}, "tail": {"text": "rule application", "start": 65, "end": 81}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "First , it uses several kinds of dictionaries to segment and tag Japanese character strings .", "output": {"relations": {"used for": [{"head": {"text": "dictionaries", "start": 33, "end": 45}, "tail": {"text": "it", "start": 8, "end": 10}}, {"head": {"text": "dictionaries", "start": 33, "end": 45}, "tail": {"text": "Japanese character strings", "start": 65, "end": 91}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Second , based on the information resulting from the dictionary lookup stage , a set of rules is applied to the segmented strings in order to identify NE items .", "output": {"relations": {"used for": [{"head": {"text": "dictionary lookup stage", "start": 53, "end": 76}, "tail": {"text": "rules", "start": 88, "end": 93}}, {"head": {"text": "rules", "start": 88, "end": 93}, "tail": {"text": "NE items", "start": 151, "end": 159}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We propose to incorporate a priori geometric constraints in a 3 -- D stereo reconstruction scheme to cope with the many cases where image information alone is not sufficient to accurately recover 3 -- D shape .", "output": {"relations": {"part of": [{"head": {"text": "priori geometric constraints", "start": 28, "end": 56}, "tail": {"text": "3 -- D stereo reconstruction scheme", "start": 62, "end": 97}}], "used for": [{"head": {"text": "image information", "start": 132, "end": 149}, "tail": {"text": "3 -- D shape", "start": 196, "end": 208}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our approach is based on the iterative deformation of a 3 -- D surface mesh to minimize an objective function .", "output": {"relations": {"used for": [{"head": {"text": "iterative deformation of a 3 -- D surface mesh", "start": 29, "end": 75}, "tail": {"text": "approach", "start": 4, "end": 12}}, {"head": {"text": "iterative deformation of a 3 -- D surface mesh", "start": 29, "end": 75}, "tail": {"text": "objective function", "start": 91, "end": 109}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We show that combining anisotropic meshing with a non-quadratic approach to regularization enables us to obtain satisfactory reconstruction results using triangulations with few vertices .", "output": {"relations": {"conjunction": [{"head": {"text": "anisotropic meshing", "start": 23, "end": 42}, "tail": {"text": "non-quadratic approach", "start": 50, "end": 72}}], "used for": [{"head": {"text": "anisotropic meshing", "start": 23, "end": 42}, "tail": {"text": "reconstruction", "start": 125, "end": 139}}, {"head": {"text": "non-quadratic approach", "start": 50, "end": 72}, "tail": {"text": "regularization", "start": 76, "end": 90}}, {"head": {"text": "non-quadratic approach", "start": 50, "end": 72}, "tail": {"text": "reconstruction", "start": 125, "end": 139}}, {"head": {"text": "triangulations", "start": 154, "end": 168}, "tail": {"text": "reconstruction", "start": 125, "end": 139}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Structural or numerical constraints can then be added locally to the reconstruction process through a constrained optimization scheme .", "output": {"relations": {"used for": [{"head": {"text": "Structural or numerical constraints", "start": 0, "end": 35}, "tail": {"text": "reconstruction process", "start": 69, "end": 91}}, {"head": {"text": "constrained optimization scheme", "start": 102, "end": 133}, "tail": {"text": "Structural or numerical constraints", "start": 0, "end": 35}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "They improve the reconstruction results and enforce their consistency with a priori knowledge about object shape .", "output": {"relations": {"used for": [{"head": {"text": "They", "start": 0, "end": 4}, "tail": {"text": "reconstruction", "start": 17, "end": 31}}], "feature of": [{"head": {"text": "object shape", "start": 100, "end": 112}, "tail": {"text": "priori knowledge", "start": 77, "end": 93}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The strong description and modeling properties of differential features make them useful tools that can be efficiently used as constraints for 3 -- D reconstruction .", "output": {"relations": {"used for": [{"head": {"text": "them", "start": 77, "end": 81}, "tail": {"text": "3 -- D reconstruction", "start": 143, "end": 164}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "It is based on a weakly supervised dependency parser that can model speech syntax without relying on any annotated training corpus .", "output": {"relations": {"used for": [{"head": {"text": "weakly supervised dependency parser", "start": 17, "end": 52}, "tail": {"text": "speech syntax", "start": 68, "end": 81}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Labeled data is replaced by a few hand-crafted rules that encode basic syntactic knowledge .", "output": {"relations": {"used for": [{"head": {"text": "hand-crafted rules", "start": 34, "end": 52}, "tail": {"text": "syntactic knowledge", "start": 71, "end": 90}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Bayesian inference then samples the rules , disambiguating and combining them to create complex tree structures that maximize a discriminative model 's posterior on a target unlabeled corpus .", "output": {"relations": {"used for": [{"head": {"text": "Bayesian inference", "start": 0, "end": 18}, "tail": {"text": "rules", "start": 36, "end": 41}}, {"head": {"text": "them", "start": 73, "end": 77}, "tail": {"text": "complex tree structures", "start": 88, "end": 111}}, {"head": {"text": "complex tree structures", "start": 88, "end": 111}, "tail": {"text": "discriminative model 's posterior", "start": 128, "end": 161}}, {"head": {"text": "unlabeled corpus", "start": 174, "end": 190}, "tail": {"text": "discriminative model 's posterior", "start": 128, "end": 161}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This posterior encodes sparse se-lectional preferences between a head word and its dependents .", "output": {"relations": {"used for": [{"head": {"text": "posterior", "start": 5, "end": 14}, "tail": {"text": "sparse se-lectional preferences", "start": 23, "end": 54}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The model is evaluated on English and Czech newspaper texts , and is then validated on French broadcast news transcriptions .", "output": {"relations": {"evaluate for": [{"head": {"text": "English and Czech newspaper texts", "start": 26, "end": 59}, "tail": {"text": "model", "start": 4, "end": 9}}, {"head": {"text": "French broadcast news transcriptions", "start": 87, "end": 123}, "tail": {"text": "model", "start": 4, "end": 9}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Listen-Communicate-Show -LRB- LCS -RRB- is a new paradigm for human interaction with data sources .", "output": {"relations": {"used for": [{"head": {"text": "Listen-Communicate-Show -LRB- LCS -RRB-", "start": 0, "end": 39}, "tail": {"text": "human interaction with data sources", "start": 62, "end": 97}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We integrate a spoken language understanding system with intelligent mobile agents that mediate between users and information sources .", "output": {"relations": {"part of": [{"head": {"text": "intelligent mobile agents", "start": 57, "end": 82}, "tail": {"text": "spoken language understanding system", "start": 15, "end": 51}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We have built and will demonstrate an application of this approach called LCS-Marine .", "output": {"relations": {"used for": [{"head": {"text": "approach", "start": 58, "end": 66}, "tail": {"text": "LCS-Marine", "start": 74, "end": 84}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A domain independent model is proposed for the automated interpretation of nominal compounds in English .", "output": {"relations": {"used for": [{"head": {"text": "domain independent model", "start": 2, "end": 26}, "tail": {"text": "automated interpretation of nominal compounds", "start": 47, "end": 92}}], "feature of": [{"head": {"text": "English", "start": 96, "end": 103}, "tail": {"text": "nominal compounds", "start": 75, "end": 92}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This model is meant to account for productive rules of interpretation which are inferred from the morpho-syntactic and semantic characteristics of the nominal constituents .", "output": {"relations": {"used for": [{"head": {"text": "model", "start": 5, "end": 10}, "tail": {"text": "productive rules of interpretation", "start": 35, "end": 69}}, {"head": {"text": "morpho-syntactic and semantic characteristics", "start": 98, "end": 143}, "tail": {"text": "productive rules of interpretation", "start": 35, "end": 69}}], "feature of": [{"head": {"text": "morpho-syntactic and semantic characteristics", "start": 98, "end": 143}, "tail": {"text": "nominal constituents", "start": 151, "end": 171}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In particular , we make extensive use of Pustejovsky 's principles concerning the predicative information associated with nominals .", "output": {"relations": {"feature of": [{"head": {"text": "nominals", "start": 122, "end": 130}, "tail": {"text": "predicative information", "start": 82, "end": 105}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We argue that it is necessary to draw a line between generalizable semantic principles and domain-specific semantic information .", "output": {"relations": {"compare": [{"head": {"text": "generalizable semantic principles", "start": 53, "end": 86}, "tail": {"text": "domain-specific semantic information", "start": 91, "end": 127}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We explain this distinction and we show how this model may be applied to the interpretation of compounds in real texts , provided that complementary semantic information are retrieved .", "output": {"relations": {"used for": [{"head": {"text": "model", "start": 49, "end": 54}, "tail": {"text": "interpretation of compounds", "start": 77, "end": 104}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We present a new method for detecting interest points using histogram information .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 17, "end": 23}, "tail": {"text": "detecting interest points", "start": 28, "end": 53}}, {"head": {"text": "histogram information", "start": 60, "end": 81}, "tail": {"text": "detecting interest points", "start": 28, "end": 53}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Unlike existing interest point detectors , which measure pixel-wise differences in image intensity , our detectors incorporate histogram-based representations , and thus can find image regions that present a distinct distribution in the neighborhood .", "output": {"relations": {"evaluate for": [{"head": {"text": "pixel-wise differences in image intensity", "start": 57, "end": 98}, "tail": {"text": "interest point detectors", "start": 16, "end": 40}}], "part of": [{"head": {"text": "histogram-based representations", "start": 127, "end": 158}, "tail": {"text": "detectors", "start": 31, "end": 40}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The proposed detectors are able to capture large-scale structures and distinctive textured patterns , and exhibit strong invariance to rotation , illumination variation , and blur .", "output": {"relations": {"used for": [{"head": {"text": "detectors", "start": 13, "end": 22}, "tail": {"text": "large-scale structures", "start": 43, "end": 65}}, {"head": {"text": "detectors", "start": 13, "end": 22}, "tail": {"text": "distinctive textured patterns", "start": 70, "end": 99}}, {"head": {"text": "detectors", "start": 13, "end": 22}, "tail": {"text": "rotation", "start": 135, "end": 143}}, {"head": {"text": "detectors", "start": 13, "end": 22}, "tail": {"text": "illumination variation", "start": 146, "end": 168}}, {"head": {"text": "detectors", "start": 13, "end": 22}, "tail": {"text": "blur", "start": 175, "end": 179}}], "conjunction": [{"head": {"text": "large-scale structures", "start": 43, "end": 65}, "tail": {"text": "distinctive textured patterns", "start": 70, "end": 99}}, {"head": {"text": "rotation", "start": 135, "end": 143}, "tail": {"text": "illumination variation", "start": 146, "end": 168}}, {"head": {"text": "illumination variation", "start": 146, "end": 168}, "tail": {"text": "blur", "start": 175, "end": 179}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The experimental results show that the proposed histogram-based interest point detectors perform particularly well for the tasks of matching textured scenes under blur and illumination changes , in terms of repeatability and distinctiveness .", "output": {"relations": {"used for": [{"head": {"text": "histogram-based interest point detectors", "start": 48, "end": 88}, "tail": {"text": "matching textured scenes", "start": 132, "end": 156}}], "evaluate for": [{"head": {"text": "repeatability", "start": 207, "end": 220}, "tail": {"text": "histogram-based interest point detectors", "start": 48, "end": 88}}, {"head": {"text": "distinctiveness", "start": 225, "end": 240}, "tail": {"text": "histogram-based interest point detectors", "start": 48, "end": 88}}], "conjunction": [{"head": {"text": "repeatability", "start": 207, "end": 220}, "tail": {"text": "distinctiveness", "start": 225, "end": 240}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "An extension of our method to space-time interest point detection for action classification is also presented .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 20, "end": 26}, "tail": {"text": "space-time interest point detection", "start": 30, "end": 65}}, {"head": {"text": "space-time interest point detection", "start": 30, "end": 65}, "tail": {"text": "action classification", "start": 70, "end": 91}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We have implemented a restricted domain parser called Plume .", "output": {"relations": {"hyponym of": [{"head": {"text": "Plume", "start": 54, "end": 59}, "tail": {"text": "restricted domain parser", "start": 22, "end": 46}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Building on previous work at Carnegie-Mellon University e.g. -LSB- 4 , 5 , 8 -RSB- , Plume 's approach to parsing is based on semantic caseframe instantiation .", "output": {"relations": {"used for": [{"head": {"text": "Plume 's approach", "start": 85, "end": 102}, "tail": {"text": "parsing", "start": 106, "end": 113}}, {"head": {"text": "semantic caseframe instantiation", "start": 126, "end": 158}, "tail": {"text": "Plume 's approach", "start": 85, "end": 102}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This has the advantages of efficiency on grammatical input , and robustness in the face of ungrammatical input .", "output": {"relations": {"feature of": [{"head": {"text": "ungrammatical input", "start": 91, "end": 110}, "tail": {"text": "robustness", "start": 65, "end": 75}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "While Plume is well adapted to simple declarative and imperative utterances , it handles passives , relative clauses and interrogatives in an ad hoc manner leading to patchy syntactic coverage .", "output": {"relations": {"used for": [{"head": {"text": "Plume", "start": 6, "end": 11}, "tail": {"text": "declarative and imperative utterances", "start": 38, "end": 75}}, {"head": {"text": "it", "start": 78, "end": 80}, "tail": {"text": "passives", "start": 89, "end": 97}}, {"head": {"text": "it", "start": 78, "end": 80}, "tail": {"text": "relative clauses", "start": 100, "end": 116}}, {"head": {"text": "it", "start": 78, "end": 80}, "tail": {"text": "interrogatives", "start": 121, "end": 135}}], "conjunction": [{"head": {"text": "passives", "start": 89, "end": 97}, "tail": {"text": "relative clauses", "start": 100, "end": 116}}, {"head": {"text": "relative clauses", "start": 100, "end": 116}, "tail": {"text": "interrogatives", "start": 121, "end": 135}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper outlines Plume as it currently exists and describes our detailed design for extending Plume to handle passives , relative clauses , and interrogatives in a general manner .", "output": {"relations": {"used for": [{"head": {"text": "Plume", "start": 20, "end": 25}, "tail": {"text": "passives", "start": 113, "end": 121}}, {"head": {"text": "Plume", "start": 20, "end": 25}, "tail": {"text": "relative clauses", "start": 124, "end": 140}}, {"head": {"text": "Plume", "start": 20, "end": 25}, "tail": {"text": "interrogatives", "start": 147, "end": 161}}], "conjunction": [{"head": {"text": "passives", "start": 113, "end": 121}, "tail": {"text": "relative clauses", "start": 124, "end": 140}}, {"head": {"text": "relative clauses", "start": 124, "end": 140}, "tail": {"text": "interrogatives", "start": 147, "end": 161}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , we present an unlexicalized parser for German which employs smoothing and suffix analysis to achieve a labelled bracket F-score of 76.2 , higher than previously reported results on the NEGRA corpus .", "output": {"relations": {"used for": [{"head": {"text": "unlexicalized parser", "start": 30, "end": 50}, "tail": {"text": "German", "start": 55, "end": 61}}, {"head": {"text": "smoothing", "start": 76, "end": 85}, "tail": {"text": "unlexicalized parser", "start": 30, "end": 50}}, {"head": {"text": "suffix analysis", "start": 90, "end": 105}, "tail": {"text": "unlexicalized parser", "start": 30, "end": 50}}], "conjunction": [{"head": {"text": "smoothing", "start": 76, "end": 85}, "tail": {"text": "suffix analysis", "start": 90, "end": 105}}], "evaluate for": [{"head": {"text": "labelled bracket F-score", "start": 119, "end": 143}, "tail": {"text": "unlexicalized parser", "start": 30, "end": 50}}, {"head": {"text": "NEGRA corpus", "start": 201, "end": 213}, "tail": {"text": "unlexicalized parser", "start": 30, "end": 50}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In addition to the high accuracy of the model , the use of smoothing in an unlexicalized parser allows us to better examine the interplay between smoothing and parsing results .", "output": {"relations": {"evaluate for": [{"head": {"text": "accuracy", "start": 24, "end": 32}, "tail": {"text": "model", "start": 40, "end": 45}}], "used for": [{"head": {"text": "smoothing", "start": 59, "end": 68}, "tail": {"text": "unlexicalized parser", "start": 75, "end": 95}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper presents an unsupervised learning approach to disambiguate various relations between named entities by use of various lexical and syntactic features from the contexts .", "output": {"relations": {"used for": [{"head": {"text": "unsupervised learning approach", "start": 23, "end": 53}, "tail": {"text": "relations between named entities", "start": 78, "end": 110}}, {"head": {"text": "lexical and syntactic features", "start": 129, "end": 159}, "tail": {"text": "unsupervised learning approach", "start": 23, "end": 53}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "It works by calculating eigenvectors of an adjacency graph 's Laplacian to recover a submanifold of data from a high dimensionality space and then performing cluster number estimation on the eigenvectors .", "output": {"relations": {"used for": [{"head": {"text": "It", "start": 0, "end": 2}, "tail": {"text": "submanifold", "start": 85, "end": 96}}, {"head": {"text": "eigenvectors", "start": 24, "end": 36}, "tail": {"text": "It", "start": 0, "end": 2}}, {"head": {"text": "high dimensionality space", "start": 112, "end": 137}, "tail": {"text": "submanifold", "start": 85, "end": 96}}, {"head": {"text": "cluster number estimation", "start": 158, "end": 183}, "tail": {"text": "It", "start": 0, "end": 2}}, {"head": {"text": "cluster number estimation", "start": 158, "end": 183}, "tail": {"text": "eigenvectors", "start": 24, "end": 36}}], "feature of": [{"head": {"text": "adjacency graph 's Laplacian", "start": 43, "end": 71}, "tail": {"text": "eigenvectors", "start": 24, "end": 36}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Experiment results on ACE corpora show that this spectral clustering based approach outperforms the other clustering methods .", "output": {"relations": {"evaluate for": [{"head": {"text": "ACE corpora", "start": 22, "end": 33}, "tail": {"text": "spectral clustering based approach", "start": 49, "end": 83}}, {"head": {"text": "ACE corpora", "start": 22, "end": 33}, "tail": {"text": "clustering methods", "start": 106, "end": 124}}], "compare": [{"head": {"text": "spectral clustering based approach", "start": 49, "end": 83}, "tail": {"text": "clustering methods", "start": 106, "end": 124}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper proposes a generic mathematical formalism for the combination of various structures : strings , trees , dags , graphs , and products of them .", "output": {"relations": {"hyponym of": [{"head": {"text": "strings", "start": 97, "end": 104}, "tail": {"text": "structures", "start": 84, "end": 94}}, {"head": {"text": "trees", "start": 107, "end": 112}, "tail": {"text": "structures", "start": 84, "end": 94}}, {"head": {"text": "dags", "start": 115, "end": 119}, "tail": {"text": "structures", "start": 84, "end": 94}}, {"head": {"text": "graphs", "start": 122, "end": 128}, "tail": {"text": "structures", "start": 84, "end": 94}}], "conjunction": [{"head": {"text": "strings", "start": 97, "end": 104}, "tail": {"text": "trees", "start": 107, "end": 112}}, {"head": {"text": "trees", "start": 107, "end": 112}, "tail": {"text": "dags", "start": 115, "end": 119}}, {"head": {"text": "dags", "start": 115, "end": 119}, "tail": {"text": "graphs", "start": 122, "end": 128}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This formalism is both elementary and powerful enough to strongly simulate many grammar formalisms , such as rewriting systems , dependency grammars , TAG , HPSG and LFG .", "output": {"relations": {"used for": [{"head": {"text": "formalism", "start": 5, "end": 14}, "tail": {"text": "grammar formalisms", "start": 80, "end": 98}}], "hyponym of": [{"head": {"text": "rewriting systems", "start": 109, "end": 126}, "tail": {"text": "grammar formalisms", "start": 80, "end": 98}}, {"head": {"text": "dependency grammars", "start": 129, "end": 148}, "tail": {"text": "grammar formalisms", "start": 80, "end": 98}}, {"head": {"text": "TAG", "start": 151, "end": 154}, "tail": {"text": "grammar formalisms", "start": 80, "end": 98}}, {"head": {"text": "HPSG", "start": 157, "end": 161}, "tail": {"text": "grammar formalisms", "start": 80, "end": 98}}, {"head": {"text": "LFG", "start": 166, "end": 169}, "tail": {"text": "grammar formalisms", "start": 80, "end": 98}}], "conjunction": [{"head": {"text": "rewriting systems", "start": 109, "end": 126}, "tail": {"text": "dependency grammars", "start": 129, "end": 148}}, {"head": {"text": "dependency grammars", "start": 129, "end": 148}, "tail": {"text": "TAG", "start": 151, "end": 154}}, {"head": {"text": "TAG", "start": 151, "end": 154}, "tail": {"text": "HPSG", "start": 157, "end": 161}}, {"head": {"text": "HPSG", "start": 157, "end": 161}, "tail": {"text": "LFG", "start": 166, "end": 169}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A mixed-signal paradigm is presented for high-resolution parallel inner-product computation in very high dimensions , suitable for efficient implementation of kernels in image processing .", "output": {"relations": {"used for": [{"head": {"text": "mixed-signal paradigm", "start": 2, "end": 23}, "tail": {"text": "high-resolution parallel inner-product computation", "start": 41, "end": 91}}, {"head": {"text": "mixed-signal paradigm", "start": 2, "end": 23}, "tail": {"text": "kernels", "start": 159, "end": 166}}, {"head": {"text": "kernels", "start": 159, "end": 166}, "tail": {"text": "image processing", "start": 170, "end": 186}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "At the core of the externally digital architecture is a high-density , low-power analog array performing binary-binary partial matrix-vector multiplication .", "output": {"relations": {"part of": [{"head": {"text": "high-density , low-power analog array", "start": 56, "end": 93}, "tail": {"text": "externally digital architecture", "start": 19, "end": 50}}], "used for": [{"head": {"text": "binary-binary partial matrix-vector multiplication", "start": 105, "end": 155}, "tail": {"text": "high-density , low-power analog array", "start": 56, "end": 93}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Full digital resolution is maintained even with low-resolution analog-to-digital conversion , owing to random statistics in the analog summation of binary products .", "output": {"relations": {"part of": [{"head": {"text": "random statistics", "start": 103, "end": 120}, "tail": {"text": "analog summation of binary products", "start": 128, "end": 163}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A random modulation scheme produces near-Bernoulli statistics even for highly correlated inputs .", "output": {"relations": {"used for": [{"head": {"text": "random modulation scheme", "start": 2, "end": 26}, "tail": {"text": "near-Bernoulli statistics", "start": 36, "end": 61}}, {"head": {"text": "highly correlated inputs", "start": 71, "end": 95}, "tail": {"text": "random modulation scheme", "start": 2, "end": 26}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The approach is validated with real image data , and with experimental results from a CID/DRAM analog array prototype in 0.5 cents m CMOS .", "output": {"relations": {"evaluate for": [{"head": {"text": "real image data", "start": 31, "end": 46}, "tail": {"text": "approach", "start": 4, "end": 12}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper we specialize the projective unifocal , bifo-cal , and trifocal tensors to the affine case , and show how the tensors obtained relate to the registered tensors encountered in previous work .", "output": {"relations": {"used for": [{"head": {"text": "projective unifocal , bifo-cal , and trifocal tensors", "start": 32, "end": 85}, "tail": {"text": "affine case", "start": 93, "end": 104}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Finally , we show how the estimation of the tensors from point correspondences is achieved through factorization , and discuss the estimation from line correspondences .", "output": {"relations": {"used for": [{"head": {"text": "point correspondences", "start": 57, "end": 78}, "tail": {"text": "estimation of the tensors", "start": 26, "end": 51}}, {"head": {"text": "factorization", "start": 99, "end": 112}, "tail": {"text": "tensors", "start": 44, "end": 51}}, {"head": {"text": "line correspondences", "start": 147, "end": 167}, "tail": {"text": "estimation", "start": 26, "end": 36}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We propose a corpus-based method -LRB- Biber ,1993 ; Nagao ,1993 ; Smadja ,1993 -RRB- which generates Noun Classifier Associations -LRB- NCA -RRB- to overcome the problems in classifier assignment and semantic construction of noun phrase .", "output": {"relations": {"used for": [{"head": {"text": "corpus-based method", "start": 13, "end": 32}, "tail": {"text": "Noun Classifier Associations -LRB- NCA -RRB-", "start": 102, "end": 146}}, {"head": {"text": "corpus-based method", "start": 13, "end": 32}, "tail": {"text": "classifier assignment", "start": 175, "end": 196}}, {"head": {"text": "corpus-based method", "start": 13, "end": 32}, "tail": {"text": "semantic construction of noun phrase", "start": 201, "end": 237}}, {"head": {"text": "Noun Classifier Associations -LRB- NCA -RRB-", "start": 102, "end": 146}, "tail": {"text": "classifier assignment", "start": 175, "end": 196}}, {"head": {"text": "Noun Classifier Associations -LRB- NCA -RRB-", "start": 102, "end": 146}, "tail": {"text": "semantic construction of noun phrase", "start": 201, "end": 237}}], "conjunction": [{"head": {"text": "classifier assignment", "start": 175, "end": 196}, "tail": {"text": "semantic construction of noun phrase", "start": 201, "end": 237}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The NCA is created statistically from a large corpus and recomposed under concept hierarchy constraints and frequency of occurrences .", "output": {"relations": {"used for": [{"head": {"text": "concept hierarchy constraints", "start": 74, "end": 103}, "tail": {"text": "NCA", "start": 4, "end": 7}}, {"head": {"text": "frequency of occurrences", "start": 108, "end": 132}, "tail": {"text": "NCA", "start": 4, "end": 7}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The perception of transparent objects from images is known to be a very hard problem in vision .", "output": {"relations": {"used for": [{"head": {"text": "images", "start": 43, "end": 49}, "tail": {"text": "perception of transparent objects", "start": 4, "end": 37}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We show how features that are imaged through a transparent object behave differently from those that are rigidly attached to the scene .", "output": {"relations": {"compare": [{"head": {"text": "those", "start": 90, "end": 95}, "tail": {"text": "features", "start": 12, "end": 20}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We present a novel model-based approach to recover the shapes and the poses of transparent objects from known motion .", "output": {"relations": {"used for": [{"head": {"text": "model-based approach", "start": 19, "end": 39}, "tail": {"text": "shapes and the poses of transparent objects", "start": 55, "end": 98}}, {"head": {"text": "known motion", "start": 104, "end": 116}, "tail": {"text": "shapes and the poses of transparent objects", "start": 55, "end": 98}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The objects can be complex in that they may be composed of multiple layers with different refractive indices .", "output": {"relations": {"part of": [{"head": {"text": "multiple layers", "start": 59, "end": 74}, "tail": {"text": "they", "start": 35, "end": 39}}], "feature of": [{"head": {"text": "refractive indices", "start": 90, "end": 108}, "tail": {"text": "multiple layers", "start": 59, "end": 74}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We have applied it to real scenes that include transparent objects and recovered the shapes of the objects with high accuracy .", "output": {"relations": {"used for": [{"head": {"text": "it", "start": 16, "end": 18}, "tail": {"text": "real scenes", "start": 22, "end": 33}}, {"head": {"text": "it", "start": 16, "end": 18}, "tail": {"text": "shapes of the objects", "start": 85, "end": 106}}], "part of": [{"head": {"text": "transparent objects", "start": 47, "end": 66}, "tail": {"text": "real scenes", "start": 22, "end": 33}}], "evaluate for": [{"head": {"text": "accuracy", "start": 117, "end": 125}, "tail": {"text": "shapes of the objects", "start": 85, "end": 106}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We propose a novel probabilistic framework for learning visual models of 3D object categories by combining appearance information and geometric constraints .", "output": {"relations": {"used for": [{"head": {"text": "probabilistic framework", "start": 19, "end": 42}, "tail": {"text": "visual models of 3D object categories", "start": 56, "end": 93}}, {"head": {"text": "appearance information", "start": 107, "end": 129}, "tail": {"text": "probabilistic framework", "start": 19, "end": 42}}, {"head": {"text": "geometric constraints", "start": 134, "end": 155}, "tail": {"text": "probabilistic framework", "start": 19, "end": 42}}], "conjunction": [{"head": {"text": "appearance information", "start": 107, "end": 129}, "tail": {"text": "geometric constraints", "start": 134, "end": 155}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A generative framework is used for learning a model that captures the relative position of parts within each of the discretized viewpoints .", "output": {"relations": {"used for": [{"head": {"text": "generative framework", "start": 2, "end": 22}, "tail": {"text": "model", "start": 46, "end": 51}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Contrary to most of the existing mixture of viewpoints models , our model establishes explicit correspondences of parts across different viewpoints of the object class .", "output": {"relations": {"compare": [{"head": {"text": "model", "start": 55, "end": 60}, "tail": {"text": "mixture of viewpoints models", "start": 33, "end": 61}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Given a new image , detection and classification are achieved by determining the position and viewpoint of the model that maximize recognition scores of the candidate objects .", "output": {"relations": {"used for": [{"head": {"text": "image", "start": 12, "end": 17}, "tail": {"text": "detection", "start": 20, "end": 29}}, {"head": {"text": "image", "start": 12, "end": 17}, "tail": {"text": "classification", "start": 34, "end": 48}}, {"head": {"text": "position", "start": 81, "end": 89}, "tail": {"text": "detection", "start": 20, "end": 29}}, {"head": {"text": "position", "start": 81, "end": 89}, "tail": {"text": "classification", "start": 34, "end": 48}}, {"head": {"text": "viewpoint", "start": 94, "end": 103}, "tail": {"text": "detection", "start": 20, "end": 29}}, {"head": {"text": "viewpoint", "start": 94, "end": 103}, "tail": {"text": "classification", "start": 34, "end": 48}}], "conjunction": [{"head": {"text": "detection", "start": 20, "end": 29}, "tail": {"text": "classification", "start": 34, "end": 48}}, {"head": {"text": "position", "start": 81, "end": 89}, "tail": {"text": "viewpoint", "start": 94, "end": 103}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our approach is among the first to propose a generative proba-bilistic framework for 3D object categorization .", "output": {"relations": {"used for": [{"head": {"text": "generative proba-bilistic framework", "start": 45, "end": 80}, "tail": {"text": "3D object categorization", "start": 85, "end": 109}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We test our algorithm on the detection task and the viewpoint classification task by using '' car '' category from both the Savarese et al. 2007 and PASCAL VOC 2006 datasets .", "output": {"relations": {"used for": [{"head": {"text": "algorithm", "start": 12, "end": 21}, "tail": {"text": "detection task", "start": 29, "end": 43}}, {"head": {"text": "algorithm", "start": 12, "end": 21}, "tail": {"text": "viewpoint classification task", "start": 52, "end": 81}}], "conjunction": [{"head": {"text": "detection task", "start": 29, "end": 43}, "tail": {"text": "viewpoint classification task", "start": 52, "end": 81}}], "evaluate for": [{"head": {"text": "PASCAL VOC 2006 datasets", "start": 149, "end": 173}, "tail": {"text": "algorithm", "start": 12, "end": 21}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We show promising results in both the detection and viewpoint classification tasks on these two challenging datasets .", "output": {"relations": {"evaluate for": [{"head": {"text": "datasets", "start": 108, "end": 116}, "tail": {"text": "detection and viewpoint classification tasks", "start": 38, "end": 82}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We present an application of ambiguity packing and stochastic disambiguation techniques for Lexical-Functional Grammars -LRB- LFG -RRB- to the domain of sentence condensation .", "output": {"relations": {"used for": [{"head": {"text": "ambiguity packing and stochastic disambiguation techniques", "start": 29, "end": 87}, "tail": {"text": "Lexical-Functional Grammars -LRB- LFG -RRB-", "start": 92, "end": 135}}, {"head": {"text": "ambiguity packing and stochastic disambiguation techniques", "start": 29, "end": 87}, "tail": {"text": "sentence condensation", "start": 153, "end": 174}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our system incorporates a linguistic parser/generator for LFG , a transfer component for parse reduction operating on packed parse forests , and a maximum-entropy model for stochastic output selection .", "output": {"relations": {"part of": [{"head": {"text": "linguistic parser/generator", "start": 26, "end": 53}, "tail": {"text": "system", "start": 4, "end": 10}}, {"head": {"text": "transfer component", "start": 66, "end": 84}, "tail": {"text": "system", "start": 4, "end": 10}}, {"head": {"text": "maximum-entropy model", "start": 147, "end": 168}, "tail": {"text": "system", "start": 4, "end": 10}}], "used for": [{"head": {"text": "linguistic parser/generator", "start": 26, "end": 53}, "tail": {"text": "LFG", "start": 58, "end": 61}}, {"head": {"text": "transfer component", "start": 66, "end": 84}, "tail": {"text": "parse reduction", "start": 89, "end": 104}}, {"head": {"text": "packed parse forests", "start": 118, "end": 138}, "tail": {"text": "parse reduction", "start": 89, "end": 104}}, {"head": {"text": "maximum-entropy model", "start": 147, "end": 168}, "tail": {"text": "stochastic output selection", "start": 173, "end": 200}}], "conjunction": [{"head": {"text": "linguistic parser/generator", "start": 26, "end": 53}, "tail": {"text": "transfer component", "start": 66, "end": 84}}, {"head": {"text": "transfer component", "start": 66, "end": 84}, "tail": {"text": "maximum-entropy model", "start": 147, "end": 168}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Furthermore , we propose the use of standard parser evaluation methods for automatically evaluating the summarization quality of sentence condensation systems .", "output": {"relations": {"evaluate for": [{"head": {"text": "parser evaluation methods", "start": 45, "end": 70}, "tail": {"text": "summarization quality", "start": 104, "end": 125}}, {"head": {"text": "summarization quality", "start": 104, "end": 125}, "tail": {"text": "sentence condensation systems", "start": 129, "end": 158}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "An experimental evaluation of summarization quality shows a close correlation between the automatic parse-based evaluation and a manual evaluation of generated strings .", "output": {"relations": {"evaluate for": [{"head": {"text": "summarization quality", "start": 30, "end": 51}, "tail": {"text": "automatic parse-based evaluation", "start": 90, "end": 122}}], "compare": [{"head": {"text": "automatic parse-based evaluation", "start": 90, "end": 122}, "tail": {"text": "manual evaluation", "start": 129, "end": 146}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Overall summarization quality of the proposed system is state-of-the-art , with guaranteed grammaticality of the system output due to the use of a constraint-based parser/generator .", "output": {"relations": {"evaluate for": [{"head": {"text": "summarization quality", "start": 8, "end": 29}, "tail": {"text": "system", "start": 46, "end": 52}}, {"head": {"text": "grammaticality", "start": 91, "end": 105}, "tail": {"text": "system", "start": 46, "end": 52}}], "used for": [{"head": {"text": "constraint-based parser/generator", "start": 147, "end": 180}, "tail": {"text": "system", "start": 46, "end": 52}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The robust principal component analysis -LRB- robust PCA -RRB- problem has been considered in many machine learning applications , where the goal is to decompose the data matrix to a low rank part plus a sparse residual .", "output": {"relations": {"used for": [{"head": {"text": "robust principal component analysis -LRB- robust PCA -RRB- problem", "start": 4, "end": 70}, "tail": {"text": "machine learning applications", "start": 99, "end": 128}}], "part of": [{"head": {"text": "low rank part", "start": 183, "end": 196}, "tail": {"text": "data matrix", "start": 166, "end": 177}}, {"head": {"text": "sparse residual", "start": 204, "end": 219}, "tail": {"text": "data matrix", "start": 166, "end": 177}}], "conjunction": [{"head": {"text": "low rank part", "start": 183, "end": 196}, "tail": {"text": "sparse residual", "start": 204, "end": 219}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "While current approaches are developed by only considering the low rank plus sparse structure , in many applications , side information of row and/or column entities may also be given , and it is still unclear to what extent could such information help robust PCA .", "output": {"relations": {"used for": [{"head": {"text": "low rank plus sparse structure", "start": 63, "end": 93}, "tail": {"text": "approaches", "start": 14, "end": 24}}, {"head": {"text": "information", "start": 124, "end": 135}, "tail": {"text": "robust PCA", "start": 253, "end": 263}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Thus , in this paper , we study the problem of robust PCA with side information , where both prior structure and features of entities are exploited for recovery .", "output": {"relations": {"used for": [{"head": {"text": "side information", "start": 63, "end": 79}, "tail": {"text": "robust PCA", "start": 47, "end": 57}}, {"head": {"text": "prior structure", "start": 93, "end": 108}, "tail": {"text": "recovery", "start": 152, "end": 160}}, {"head": {"text": "features of entities", "start": 113, "end": 133}, "tail": {"text": "recovery", "start": 152, "end": 160}}], "conjunction": [{"head": {"text": "prior structure", "start": 93, "end": 108}, "tail": {"text": "features of entities", "start": 113, "end": 133}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We propose a convex problem to incorporate side information in robust PCA and show that the low rank matrix can be exactly recovered via the proposed method under certain conditions .", "output": {"relations": {"used for": [{"head": {"text": "convex problem", "start": 13, "end": 27}, "tail": {"text": "side information", "start": 43, "end": 59}}, {"head": {"text": "method", "start": 150, "end": 156}, "tail": {"text": "low rank matrix", "start": 92, "end": 107}}], "part of": [{"head": {"text": "side information", "start": 43, "end": 59}, "tail": {"text": "robust PCA", "start": 63, "end": 73}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In particular , our guarantee suggests that a substantial amount of low rank matrices , which can not be recovered by standard robust PCA , become re-coverable by our proposed method .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 176, "end": 182}, "tail": {"text": "low rank matrices", "start": 68, "end": 85}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The result theoretically justifies the effectiveness of features in robust PCA .", "output": {"relations": {"feature of": [{"head": {"text": "features", "start": 56, "end": 64}, "tail": {"text": "robust PCA", "start": 68, "end": 78}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In addition , we conduct synthetic experiments as well as a real application on noisy image classification to show that our method also improves the performance in practice by exploiting side information .", "output": {"relations": {"evaluate for": [{"head": {"text": "noisy image classification", "start": 80, "end": 106}, "tail": {"text": "method", "start": 124, "end": 130}}], "used for": [{"head": {"text": "side information", "start": 187, "end": 203}, "tail": {"text": "method", "start": 124, "end": 130}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper presents necessary and sufficient conditions for the use of demonstrative expressions in English and discusses implications for current discourse processing algorithms .", "output": {"relations": {"feature of": [{"head": {"text": "demonstrative expressions", "start": 71, "end": 96}, "tail": {"text": "English", "start": 100, "end": 107}}], "used for": [{"head": {"text": "implications", "start": 122, "end": 134}, "tail": {"text": "discourse processing algorithms", "start": 147, "end": 178}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This research is part of a larger study of anaphoric expressions , the results of which will be incorporated into a natural language generation system .", "output": {"relations": {"used for": [{"head": {"text": "anaphoric expressions", "start": 43, "end": 64}, "tail": {"text": "natural language generation system", "start": 116, "end": 150}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Using the IEMOCAP database , discrete -LRB- categorical -RRB- and continuous -LRB- attribute -RRB- emotional assessments evaluated by the actors and na ¨ ıve listeners are compared .", "output": {"relations": {"used for": [{"head": {"text": "IEMOCAP database", "start": 10, "end": 26}, "tail": {"text": "discrete -LRB- categorical -RRB- and continuous -LRB- attribute -RRB- emotional assessments", "start": 29, "end": 120}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The problem of blind separation of underdetermined instantaneous mixtures of independent signals is addressed through a method relying on nonstationarity of the original signals .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 120, "end": 126}, "tail": {"text": "blind separation of underdetermined instantaneous mixtures of independent signals", "start": 15, "end": 96}}, {"head": {"text": "nonstationarity", "start": 138, "end": 153}, "tail": {"text": "method", "start": 120, "end": 126}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In comparison with previous works , in this paper it is assumed that the signals are not i.i.d. in each epoch , but obey a first-order autoregressive model .", "output": {"relations": {"used for": [{"head": {"text": "first-order autoregressive model", "start": 123, "end": 155}, "tail": {"text": "signals", "start": 73, "end": 80}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This model was shown to be more appropriate for blind separation of natural speech signals .", "output": {"relations": {"used for": [{"head": {"text": "model", "start": 5, "end": 10}, "tail": {"text": "blind separation of natural speech signals .", "start": 48, "end": 92}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A separation method is proposed that is nearly statistically efficient -LRB- approaching the corresponding Cramér-Rao lower bound -RRB- , if the separated signals obey the assumed model .", "output": {"relations": {"feature of": [{"head": {"text": "Cramér-Rao lower bound -RRB-", "start": 107, "end": 135}, "tail": {"text": "separation method", "start": 2, "end": 19}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In the case of natural speech signals , the method is shown to have separation accuracy better than the state-of-the-art methods .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 44, "end": 50}, "tail": {"text": "natural speech signals", "start": 15, "end": 37}}, {"head": {"text": "methods", "start": 121, "end": 128}, "tail": {"text": "natural speech signals", "start": 15, "end": 37}}], "compare": [{"head": {"text": "method", "start": 44, "end": 50}, "tail": {"text": "methods", "start": 121, "end": 128}}], "evaluate for": [{"head": {"text": "separation accuracy", "start": 68, "end": 87}, "tail": {"text": "method", "start": 44, "end": 50}}, {"head": {"text": "separation accuracy", "start": 68, "end": 87}, "tail": {"text": "methods", "start": 121, "end": 128}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper proposes to use a convolution kernel over parse trees to model syntactic structure information for relation extraction .", "output": {"relations": {"used for": [{"head": {"text": "convolution kernel over parse trees", "start": 29, "end": 64}, "tail": {"text": "syntactic structure information", "start": 74, "end": 105}}, {"head": {"text": "syntactic structure information", "start": 74, "end": 105}, "tail": {"text": "relation extraction", "start": 110, "end": 129}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our study reveals that the syntactic structure features embedded in a parse tree are very effective for relation extraction and these features can be well captured by the convolution tree kernel .", "output": {"relations": {"feature of": [{"head": {"text": "syntactic structure features", "start": 27, "end": 55}, "tail": {"text": "parse tree", "start": 70, "end": 80}}], "used for": [{"head": {"text": "syntactic structure features", "start": 27, "end": 55}, "tail": {"text": "relation extraction", "start": 104, "end": 123}}, {"head": {"text": "convolution tree kernel", "start": 171, "end": 194}, "tail": {"text": "features", "start": 47, "end": 55}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Evaluation on the ACE 2003 corpus shows that the convolution kernel over parse trees can achieve comparable performance with the previous best-reported feature-based methods on the 24 ACE relation subtypes .", "output": {"relations": {"evaluate for": [{"head": {"text": "ACE 2003 corpus", "start": 18, "end": 33}, "tail": {"text": "convolution kernel over parse trees", "start": 49, "end": 84}}], "compare": [{"head": {"text": "feature-based methods", "start": 152, "end": 173}, "tail": {"text": "convolution kernel over parse trees", "start": 49, "end": 84}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "It also shows that our method significantly outperforms the previous two dependency tree kernels on the 5 ACE relation major types .", "output": {"relations": {"compare": [{"head": {"text": "method", "start": 23, "end": 29}, "tail": {"text": "dependency tree kernels", "start": 73, "end": 96}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper presents the results of automatically inducing a Combinatory Categorial Grammar -LRB- CCG -RRB- lexicon from a Turkish dependency treebank .", "output": {"relations": {"part of": [{"head": {"text": "Combinatory Categorial Grammar -LRB- CCG -RRB- lexicon", "start": 60, "end": 114}, "tail": {"text": "Turkish dependency treebank", "start": 122, "end": 149}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The fact that Turkish is an agglutinating free word order language presents a challenge for language theories .", "output": {"relations": {"hyponym of": [{"head": {"text": "Turkish", "start": 14, "end": 21}, "tail": {"text": "agglutinating free word order language", "start": 28, "end": 66}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We explored possible ways to obtain a compact lexicon , consistent with CCG principles , from a treebank which is an order of magnitude smaller than Penn WSJ .", "output": {"relations": {"part of": [{"head": {"text": "compact lexicon", "start": 38, "end": 53}, "tail": {"text": "treebank", "start": 96, "end": 104}}], "compare": [{"head": {"text": "treebank", "start": 96, "end": 104}, "tail": {"text": "Penn WSJ", "start": 149, "end": 157}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "While sentence extraction as an approach to summarization has been shown to work in documents of certain genres , because of the conversational nature of email communication where utterances are made in relation to one made previously , sentence extraction may not capture the necessary segments of dialogue that would make a summary coherent .", "output": {"relations": {"used for": [{"head": {"text": "sentence extraction", "start": 6, "end": 25}, "tail": {"text": "summarization", "start": 44, "end": 57}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , we present our work on the detection of question-answer pairs in an email conversation for the task of email summarization .", "output": {"relations": {"used for": [{"head": {"text": "detection of question-answer pairs", "start": 43, "end": 77}, "tail": {"text": "email summarization", "start": 119, "end": 138}}, {"head": {"text": "email conversation", "start": 84, "end": 102}, "tail": {"text": "detection of question-answer pairs", "start": 43, "end": 77}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We show that various features based on the structure of email-threads can be used to improve upon lexical similarity of discourse segments for question-answer pairing .", "output": {"relations": {"used for": [{"head": {"text": "features", "start": 21, "end": 29}, "tail": {"text": "lexical similarity", "start": 98, "end": 116}}, {"head": {"text": "features", "start": 21, "end": 29}, "tail": {"text": "question-answer pairing", "start": 143, "end": 166}}, {"head": {"text": "structure of email-threads", "start": 43, "end": 69}, "tail": {"text": "features", "start": 21, "end": 29}}], "feature of": [{"head": {"text": "lexical similarity", "start": 98, "end": 116}, "tail": {"text": "discourse segments", "start": 120, "end": 138}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Specifically , we show how to incorporate a simple prior on the distribution of natural images into support vector machines .", "output": {"relations": {"used for": [{"head": {"text": "prior on the distribution of natural images", "start": 51, "end": 94}, "tail": {"text": "support vector machines", "start": 100, "end": 123}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "SVMs are known to be robust to overfitting ; however , a few training examples usually do not represent well the structure of the class .", "output": {"relations": {"used for": [{"head": {"text": "SVMs", "start": 0, "end": 4}, "tail": {"text": "overfitting", "start": 31, "end": 42}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our experiments on real data sets show that the resulting detector is more robust to the choice of training examples , and substantially improves both linear and kernel SVM when trained on 10 positive and 10 negative examples .", "output": {"relations": {"evaluate for": [{"head": {"text": "real data sets", "start": 19, "end": 33}, "tail": {"text": "detector", "start": 58, "end": 66}}], "compare": [{"head": {"text": "detector", "start": 58, "end": 66}, "tail": {"text": "linear and kernel SVM", "start": 151, "end": 172}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Although the study of clustering is centered around an intuitively compelling goal , it has been very difficult to develop a unified framework for reasoning about it at a technical level , and profoundly diverse approaches to clustering abound in the research community .", "output": {"relations": {"used for": [{"head": {"text": "unified framework", "start": 125, "end": 142}, "tail": {"text": "reasoning", "start": 147, "end": 156}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Relaxations of these properties expose some of the interesting -LRB- and unavoidable -RRB- trade-offs at work in well-studied clustering techniques such as single-linkage , sum-of-pairs , k-means , and k-median .", "output": {"relations": {"hyponym of": [{"head": {"text": "single-linkage", "start": 156, "end": 170}, "tail": {"text": "well-studied clustering techniques", "start": 113, "end": 147}}, {"head": {"text": "sum-of-pairs", "start": 173, "end": 185}, "tail": {"text": "well-studied clustering techniques", "start": 113, "end": 147}}, {"head": {"text": "k-means", "start": 188, "end": 195}, "tail": {"text": "well-studied clustering techniques", "start": 113, "end": 147}}, {"head": {"text": "k-median", "start": 202, "end": 210}, "tail": {"text": "well-studied clustering techniques", "start": 113, "end": 147}}], "conjunction": [{"head": {"text": "single-linkage", "start": 156, "end": 170}, "tail": {"text": "sum-of-pairs", "start": 173, "end": 185}}, {"head": {"text": "sum-of-pairs", "start": 173, "end": 185}, "tail": {"text": "k-means", "start": 188, "end": 195}}, {"head": {"text": "k-means", "start": 188, "end": 195}, "tail": {"text": "k-median", "start": 202, "end": 210}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "With relevant approach , we identify important contents by PageRank algorithm on the event map constructed from documents .", "output": {"relations": {"used for": [{"head": {"text": "PageRank algorithm", "start": 59, "end": 77}, "tail": {"text": "relevant approach", "start": 5, "end": 22}}, {"head": {"text": "event map", "start": 85, "end": 94}, "tail": {"text": "PageRank algorithm", "start": 59, "end": 77}}, {"head": {"text": "documents", "start": 112, "end": 121}, "tail": {"text": "event map", "start": 85, "end": 94}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We present a scanning method that recovers dense sub-pixel camera-projector correspondence without requiring any photometric calibration nor preliminary knowledge of their relative geometry .", "output": {"relations": {"used for": [{"head": {"text": "scanning method", "start": 13, "end": 28}, "tail": {"text": "dense sub-pixel camera-projector correspondence", "start": 43, "end": 90}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Subpixel accuracy is achieved by considering several zero-crossings defined by the difference between pairs of unstructured patterns .", "output": {"relations": {"used for": [{"head": {"text": "zero-crossings", "start": 53, "end": 67}, "tail": {"text": "Subpixel accuracy", "start": 0, "end": 17}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We use gray-level band-pass white noise patterns that increase robustness to indirect lighting and scene discontinuities .", "output": {"relations": {"evaluate for": [{"head": {"text": "robustness", "start": 63, "end": 73}, "tail": {"text": "gray-level band-pass white noise patterns", "start": 7, "end": 48}}], "feature of": [{"head": {"text": "indirect lighting", "start": 77, "end": 94}, "tail": {"text": "robustness", "start": 63, "end": 73}}, {"head": {"text": "scene discontinuities", "start": 99, "end": 120}, "tail": {"text": "robustness", "start": 63, "end": 73}}], "conjunction": [{"head": {"text": "indirect lighting", "start": 77, "end": 94}, "tail": {"text": "scene discontinuities", "start": 99, "end": 120}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Simulated and experimental results show that our method recovers scene geometry with high subpixel precision , and that it can handle many challenges of active reconstruction systems .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 49, "end": 55}, "tail": {"text": "scene geometry", "start": 65, "end": 79}}, {"head": {"text": "it", "start": 81, "end": 83}, "tail": {"text": "active reconstruction systems", "start": 153, "end": 182}}], "feature of": [{"head": {"text": "subpixel precision", "start": 90, "end": 108}, "tail": {"text": "scene geometry", "start": 65, "end": 79}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We compare our results to state of the art methods such as mi-cro phase shifting and modulated phase shifting .", "output": {"relations": {"hyponym of": [{"head": {"text": "mi-cro phase shifting", "start": 59, "end": 80}, "tail": {"text": "state of the art methods", "start": 26, "end": 50}}, {"head": {"text": "modulated phase shifting", "start": 85, "end": 109}, "tail": {"text": "state of the art methods", "start": 26, "end": 50}}], "conjunction": [{"head": {"text": "mi-cro phase shifting", "start": 59, "end": 80}, "tail": {"text": "modulated phase shifting", "start": 85, "end": 109}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper describes a novel system for acquiring adjectival subcategorization frames -LRB- scfs -RRB- and associated frequency information from English corpus data .", "output": {"relations": {"used for": [{"head": {"text": "system", "start": 29, "end": 35}, "tail": {"text": "acquiring adjectival subcategorization frames -LRB- scfs -RRB-", "start": 40, "end": 102}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The system incorporates a decision-tree classifier for 30 scf types which tests for the presence of grammatical relations -LRB- grs -RRB- in the output of a robust statistical parser .", "output": {"relations": {"part of": [{"head": {"text": "decision-tree classifier", "start": 26, "end": 50}, "tail": {"text": "system", "start": 4, "end": 10}}], "used for": [{"head": {"text": "decision-tree classifier", "start": 26, "end": 50}, "tail": {"text": "grammatical relations -LRB- grs -RRB-", "start": 100, "end": 137}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "It uses a powerful pattern-matching language to classify grs into frames hierarchically in a way that mirrors inheritance-based lexica .", "output": {"relations": {"used for": [{"head": {"text": "pattern-matching language", "start": 19, "end": 44}, "tail": {"text": "It", "start": 0, "end": 2}}, {"head": {"text": "pattern-matching language", "start": 19, "end": 44}, "tail": {"text": "grs", "start": 57, "end": 60}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The experiments show that the system is able to detect scf types with 70 % precision and 66 % recall rate .", "output": {"relations": {"evaluate for": [{"head": {"text": "precision", "start": 75, "end": 84}, "tail": {"text": "system", "start": 30, "end": 36}}, {"head": {"text": "recall", "start": 94, "end": 100}, "tail": {"text": "system", "start": 30, "end": 36}}], "conjunction": [{"head": {"text": "precision", "start": 75, "end": 84}, "tail": {"text": "recall", "start": 94, "end": 100}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A new tool for linguistic annotation of scfs in corpus data is also introduced which can considerably alleviate the process of obtaining training and test data for subcategorization acquisition .", "output": {"relations": {"used for": [{"head": {"text": "tool", "start": 6, "end": 10}, "tail": {"text": "linguistic annotation of scfs", "start": 15, "end": 44}}, {"head": {"text": "training and test data", "start": 137, "end": 159}, "tail": {"text": "subcategorization acquisition", "start": 164, "end": 193}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Machine transliteration/back-transliteration plays an important role in many multilingual speech and language applications .", "output": {"relations": {"used for": [{"head": {"text": "Machine transliteration/back-transliteration", "start": 0, "end": 44}, "tail": {"text": "multilingual speech and language applications", "start": 77, "end": 122}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , a novel framework for machine transliteration/backtransliteration that allows us to carry out direct orthographical mapping -LRB- DOM -RRB- between two different languages is presented .", "output": {"relations": {"used for": [{"head": {"text": "framework", "start": 24, "end": 33}, "tail": {"text": "machine transliteration/backtransliteration", "start": 38, "end": 81}}, {"head": {"text": "machine transliteration/backtransliteration", "start": 38, "end": 81}, "tail": {"text": "direct orthographical mapping -LRB- DOM -RRB-", "start": 110, "end": 155}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Under this framework , a joint source-channel transliteration model , also called n-gram transliteration model -LRB- n-gram TM -RRB- , is further proposed to model the transliteration process .", "output": {"relations": {"used for": [{"head": {"text": "framework", "start": 11, "end": 20}, "tail": {"text": "joint source-channel transliteration model", "start": 25, "end": 67}}, {"head": {"text": "n-gram transliteration model -LRB- n-gram TM -RRB-", "start": 82, "end": 132}, "tail": {"text": "transliteration process", "start": 168, "end": 191}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We evaluate the proposed methods through several transliteration/backtransliteration experiments for English/Chinese and English/Japanese language pairs .", "output": {"relations": {"evaluate for": [{"head": {"text": "transliteration/backtransliteration", "start": 49, "end": 84}, "tail": {"text": "methods", "start": 25, "end": 32}}], "used for": [{"head": {"text": "transliteration/backtransliteration", "start": 49, "end": 84}, "tail": {"text": "English/Chinese and English/Japanese language pairs", "start": 101, "end": 152}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our study reveals that the proposed method not only reduces an extensive system development effort but also improves the transliteration accuracy significantly .", "output": {"relations": {"evaluate for": [{"head": {"text": "transliteration accuracy", "start": 121, "end": 145}, "tail": {"text": "method", "start": 36, "end": 42}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A bio-inspired model for an analog programmable array processor -LRB- APAP -RRB- , based on studies on the vertebrate retina , has permitted the realization of complex programmable spatio-temporal dynamics in VLSI .", "output": {"relations": {"used for": [{"head": {"text": "bio-inspired model", "start": 2, "end": 20}, "tail": {"text": "analog programmable array processor -LRB- APAP -RRB-", "start": 28, "end": 80}}, {"head": {"text": "bio-inspired model", "start": 2, "end": 20}, "tail": {"text": "complex programmable spatio-temporal dynamics", "start": 160, "end": 205}}, {"head": {"text": "vertebrate retina", "start": 107, "end": 124}, "tail": {"text": "bio-inspired model", "start": 2, "end": 20}}], "feature of": [{"head": {"text": "complex programmable spatio-temporal dynamics", "start": 160, "end": 205}, "tail": {"text": "VLSI", "start": 209, "end": 213}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This model mimics the way in which images are processed in the visual pathway , rendering a feasible alternative for the implementation of early vision applications in standard technologies .", "output": {"relations": {"used for": [{"head": {"text": "visual pathway", "start": 63, "end": 77}, "tail": {"text": "images", "start": 35, "end": 41}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Computing power per area and power consumption is amongst the highest reported for a single chip .", "output": {"relations": {"conjunction": [{"head": {"text": "Computing power per area", "start": 0, "end": 24}, "tail": {"text": "power consumption", "start": 29, "end": 46}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Another problem with determiners is their inherent ambiguity .", "output": {"relations": {"feature of": [{"head": {"text": "ambiguity", "start": 51, "end": 60}, "tail": {"text": "determiners", "start": 21, "end": 32}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper we propose a logical formalism , which , among other things , is suitable for representing determiners without forcing a particular interpretation when their meaning is still not clear .", "output": {"relations": {"used for": [{"head": {"text": "logical formalism", "start": 27, "end": 44}, "tail": {"text": "determiners", "start": 105, "end": 116}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We investigate the verbal and nonverbal means for grounding , and propose a design for embodied conversational agents that relies on both kinds of signals to establish common ground in human-computer interaction .", "output": {"relations": {"used for": [{"head": {"text": "verbal and nonverbal means", "start": 19, "end": 45}, "tail": {"text": "grounding", "start": 50, "end": 59}}, {"head": {"text": "design", "start": 76, "end": 82}, "tail": {"text": "embodied conversational agents", "start": 87, "end": 117}}, {"head": {"text": "common ground", "start": 168, "end": 181}, "tail": {"text": "human-computer interaction", "start": 185, "end": 211}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We analyzed eye gaze , head nods and attentional focus in the context of a direction-giving task .", "output": {"relations": {"conjunction": [{"head": {"text": "eye gaze", "start": 12, "end": 20}, "tail": {"text": "head nods", "start": 23, "end": 32}}, {"head": {"text": "head nods", "start": 23, "end": 32}, "tail": {"text": "attentional focus", "start": 37, "end": 54}}], "part of": [{"head": {"text": "eye gaze", "start": 12, "end": 20}, "tail": {"text": "direction-giving task", "start": 75, "end": 96}}, {"head": {"text": "head nods", "start": 23, "end": 32}, "tail": {"text": "direction-giving task", "start": 75, "end": 96}}, {"head": {"text": "attentional focus", "start": 37, "end": 54}, "tail": {"text": "direction-giving task", "start": 75, "end": 96}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Based on these results , we present an ECA that uses verbal and nonverbal grounding acts to update dialogue state .", "output": {"relations": {"used for": [{"head": {"text": "verbal and nonverbal grounding acts", "start": 53, "end": 88}, "tail": {"text": "ECA", "start": 39, "end": 42}}, {"head": {"text": "verbal and nonverbal grounding acts", "start": 53, "end": 88}, "tail": {"text": "dialogue state", "start": 99, "end": 113}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Sentence boundary detection in speech is important for enriching speech recognition output , making it easier for humans to read and downstream modules to process .", "output": {"relations": {"used for": [{"head": {"text": "Sentence boundary detection", "start": 0, "end": 27}, "tail": {"text": "speech recognition output", "start": 65, "end": 90}}, {"head": {"text": "speech", "start": 31, "end": 37}, "tail": {"text": "Sentence boundary detection", "start": 0, "end": 27}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In previous work , we have developed hidden Markov model -LRB- HMM -RRB- and maximum entropy -LRB- Maxent -RRB- classifiers that integrate textual and prosodic knowledge sources for detecting sentence boundaries .", "output": {"relations": {"used for": [{"head": {"text": "hidden Markov model -LRB- HMM -RRB- and maximum entropy -LRB- Maxent -RRB- classifiers", "start": 37, "end": 123}, "tail": {"text": "detecting sentence boundaries", "start": 182, "end": 211}}, {"head": {"text": "textual and prosodic knowledge sources", "start": 139, "end": 177}, "tail": {"text": "hidden Markov model -LRB- HMM -RRB- and maximum entropy -LRB- Maxent -RRB- classifiers", "start": 37, "end": 123}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , we evaluate the use of a conditional random field -LRB- CRF -RRB- for this task and relate results with this model to our prior work .", "output": {"relations": {"used for": [{"head": {"text": "conditional random field -LRB- CRF -RRB-", "start": 41, "end": 81}, "tail": {"text": "task", "start": 91, "end": 95}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We evaluate across two corpora -LRB- conversational telephone speech and broadcast news speech -RRB- on both human transcriptions and speech recognition output .", "output": {"relations": {"evaluate for": [{"head": {"text": "corpora", "start": 23, "end": 30}, "tail": {"text": "human transcriptions", "start": 109, "end": 129}}, {"head": {"text": "corpora", "start": 23, "end": 30}, "tail": {"text": "speech recognition output", "start": 134, "end": 159}}], "hyponym of": [{"head": {"text": "conversational telephone speech", "start": 37, "end": 68}, "tail": {"text": "corpora", "start": 23, "end": 30}}, {"head": {"text": "broadcast news speech", "start": 73, "end": 94}, "tail": {"text": "corpora", "start": 23, "end": 30}}], "conjunction": [{"head": {"text": "conversational telephone speech", "start": 37, "end": 68}, "tail": {"text": "broadcast news speech", "start": 73, "end": 94}}, {"head": {"text": "human transcriptions", "start": 109, "end": 129}, "tail": {"text": "speech recognition output", "start": 134, "end": 159}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In general , our CRF model yields a lower error rate than the HMM and Max-ent models on the NIST sentence boundary detection task in speech , although it is interesting to note that the best results are achieved by three-way voting among the classifiers .", "output": {"relations": {"compare": [{"head": {"text": "CRF model", "start": 17, "end": 26}, "tail": {"text": "HMM and Max-ent models", "start": 62, "end": 84}}], "evaluate for": [{"head": {"text": "error rate", "start": 42, "end": 52}, "tail": {"text": "CRF model", "start": 17, "end": 26}}, {"head": {"text": "error rate", "start": 42, "end": 52}, "tail": {"text": "HMM and Max-ent models", "start": 62, "end": 84}}, {"head": {"text": "NIST sentence boundary detection task", "start": 92, "end": 129}, "tail": {"text": "CRF model", "start": 17, "end": 26}}, {"head": {"text": "NIST sentence boundary detection task", "start": 92, "end": 129}, "tail": {"text": "HMM and Max-ent models", "start": 62, "end": 84}}], "feature of": [{"head": {"text": "speech", "start": 133, "end": 139}, "tail": {"text": "NIST sentence boundary detection task", "start": 92, "end": 129}}], "used for": [{"head": {"text": "classifiers", "start": 242, "end": 253}, "tail": {"text": "three-way voting", "start": 215, "end": 231}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This probably occurs because each model has different strengths and weaknesses for modeling the knowledge sources .", "output": {"relations": {"used for": [{"head": {"text": "model", "start": 34, "end": 39}, "tail": {"text": "knowledge sources", "start": 96, "end": 113}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We propose a novel approach to associate objects across multiple PTZ cameras that can be used to perform camera handoff in wide-area surveillance scenarios .", "output": {"relations": {"used for": [{"head": {"text": "approach", "start": 19, "end": 27}, "tail": {"text": "camera handoff in wide-area surveillance scenarios", "start": 105, "end": 155}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "While previous approaches relied on geometric , appearance , or correlation-based information for establishing correspondences between static cameras , they each have well-known limitations and are not extendable to wide-area settings with PTZ cameras .", "output": {"relations": {"used for": [{"head": {"text": "geometric , appearance , or correlation-based information", "start": 36, "end": 93}, "tail": {"text": "approaches", "start": 15, "end": 25}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Towards this goal , we also propose a novel Multiple Instance Learning -LRB- MIL -RRB- formulation for the problem based on the logistic softmax function of covariance-based region features within a MAP estimation framework .", "output": {"relations": {"used for": [{"head": {"text": "logistic softmax function of covariance-based region features", "start": 128, "end": 189}, "tail": {"text": "Multiple Instance Learning -LRB- MIL -RRB- formulation", "start": 44, "end": 98}}, {"head": {"text": "MAP estimation framework", "start": 199, "end": 223}, "tail": {"text": "Multiple Instance Learning -LRB- MIL -RRB- formulation", "start": 44, "end": 98}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We demonstrate our approach with multiple PTZ camera sequences in typical outdoor surveillance settings and show a comparison with state-of-the-art approaches .", "output": {"relations": {"used for": [{"head": {"text": "approach", "start": 19, "end": 27}, "tail": {"text": "outdoor surveillance settings", "start": 74, "end": 103}}, {"head": {"text": "multiple PTZ camera sequences", "start": 33, "end": 62}, "tail": {"text": "approach", "start": 19, "end": 27}}], "compare": [{"head": {"text": "approach", "start": 19, "end": 27}, "tail": {"text": "state-of-the-art approaches", "start": 131, "end": 158}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper solves a specialized regression problem to obtain sampling probabilities for records in databases .", "output": {"relations": {"used for": [{"head": {"text": "specialized regression problem", "start": 20, "end": 50}, "tail": {"text": "sampling probabilities", "start": 61, "end": 83}}, {"head": {"text": "sampling probabilities", "start": 61, "end": 83}, "tail": {"text": "records", "start": 88, "end": 95}}], "part of": [{"head": {"text": "records", "start": 88, "end": 95}, "tail": {"text": "databases", "start": 99, "end": 108}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The goal is to sample a small set of records over which evaluating aggregate queries can be done both efficiently and accurately .", "output": {"relations": {"evaluate for": [{"head": {"text": "aggregate queries", "start": 67, "end": 84}, "tail": {"text": "records", "start": 37, "end": 44}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We provide a principled and provable solution for this problem ; it is parameterless and requires no data insights .", "output": {"relations": {"used for": [{"head": {"text": "principled and provable solution", "start": 13, "end": 45}, "tail": {"text": "problem", "start": 55, "end": 62}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Moreover , a cost zero solution always exists and can only be excluded by hard budget constraints .", "output": {"relations": {"used for": [{"head": {"text": "hard budget constraints", "start": 74, "end": 97}, "tail": {"text": "cost zero solution", "start": 13, "end": 31}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our extensive experimental results significantly improve over both uniform sampling and standard stratified sampling which are de-facto the industry standards .", "output": {"relations": {"conjunction": [{"head": {"text": "uniform sampling", "start": 67, "end": 83}, "tail": {"text": "stratified sampling", "start": 97, "end": 116}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We consider the problem of computing the Kullback-Leibler distance , also called the relative entropy , between a probabilistic context-free grammar and a probabilistic finite automaton .", "output": {"relations": {"compare": [{"head": {"text": "probabilistic context-free grammar", "start": 114, "end": 148}, "tail": {"text": "probabilistic finite automaton", "start": 155, "end": 185}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We show that there is a closed-form -LRB- analytical -RRB- solution for one part of the Kullback-Leibler distance , viz the cross-entropy .", "output": {"relations": {"used for": [{"head": {"text": "closed-form -LRB- analytical -RRB- solution", "start": 24, "end": 67}, "tail": {"text": "Kullback-Leibler distance", "start": 88, "end": 113}}, {"head": {"text": "closed-form -LRB- analytical -RRB- solution", "start": 24, "end": 67}, "tail": {"text": "cross-entropy", "start": 124, "end": 137}}], "part of": [{"head": {"text": "cross-entropy", "start": 124, "end": 137}, "tail": {"text": "Kullback-Leibler distance", "start": 88, "end": 113}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We discuss several applications of the result to the problem of distributional approximation of probabilistic context-free grammars by means of probabilistic finite automata .", "output": {"relations": {"feature of": [{"head": {"text": "distributional approximation", "start": 64, "end": 92}, "tail": {"text": "probabilistic context-free grammars", "start": 96, "end": 131}}], "used for": [{"head": {"text": "probabilistic finite automata", "start": 144, "end": 173}, "tail": {"text": "distributional approximation", "start": 64, "end": 92}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In spite of over two decades of intense research , illumination and pose invariance remain prohibitively challenging aspects of face recognition for most practical applications .", "output": {"relations": {"conjunction": [{"head": {"text": "illumination", "start": 51, "end": 63}, "tail": {"text": "pose invariance", "start": 68, "end": 83}}], "part of": [{"head": {"text": "illumination", "start": 51, "end": 63}, "tail": {"text": "face recognition", "start": 128, "end": 144}}, {"head": {"text": "pose invariance", "start": 68, "end": 83}, "tail": {"text": "face recognition", "start": 128, "end": 144}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The objective of this work is to recognize faces using video sequences both for training and recognition input , in a realistic , unconstrained setup in which lighting , pose and user motion pattern have a wide variability and face images are of low resolution .", "output": {"relations": {"conjunction": [{"head": {"text": "lighting", "start": 159, "end": 167}, "tail": {"text": "pose", "start": 170, "end": 174}}, {"head": {"text": "pose", "start": 170, "end": 174}, "tail": {"text": "user motion pattern", "start": 179, "end": 198}}], "feature of": [{"head": {"text": "resolution", "start": 250, "end": 260}, "tail": {"text": "face images", "start": 227, "end": 238}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In particular there are three areas of novelty : -LRB- i -RRB- we show how a photometric model of image formation can be combined with a statistical model of generic face appearance variation , learnt offline , to generalize in the presence of extreme illumination changes ; -LRB- ii -RRB- we use the smoothness of geodesically local appearance manifold structure and a robust same-identity likelihood to achieve invariance to unseen head poses ; and -LRB- iii -RRB- we introduce an accurate video sequence '' reillumination '' algorithm to achieve robustness to face motion patterns in video .", "output": {"relations": {"used for": [{"head": {"text": "photometric model", "start": 77, "end": 94}, "tail": {"text": "image formation", "start": 98, "end": 113}}, {"head": {"text": "statistical model", "start": 137, "end": 154}, "tail": {"text": "generic face appearance variation", "start": 158, "end": 191}}, {"head": {"text": "statistical model", "start": 137, "end": 154}, "tail": {"text": "extreme illumination changes", "start": 244, "end": 272}}], "conjunction": [{"head": {"text": "photometric model", "start": 77, "end": 94}, "tail": {"text": "statistical model", "start": 137, "end": 154}}, {"head": {"text": "geodesically local appearance manifold structure", "start": 315, "end": 363}, "tail": {"text": "robust same-identity likelihood", "start": 370, "end": 401}}], "feature of": [{"head": {"text": "smoothness", "start": 301, "end": 311}, "tail": {"text": "geodesically local appearance manifold structure", "start": 315, "end": 363}}, {"head": {"text": "face motion patterns", "start": 563, "end": 583}, "tail": {"text": "robustness", "start": 549, "end": 559}}], "evaluate for": [{"head": {"text": "robustness", "start": 549, "end": 559}, "tail": {"text": "video sequence '' reillumination '' algorithm", "start": 492, "end": 537}}], "part of": [{"head": {"text": "face motion patterns", "start": 563, "end": 583}, "tail": {"text": "video", "start": 492, "end": 497}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We describe a fully automatic recognition system based on the proposed method and an extensive evaluation on 171 individuals and over 1300 video sequences with extreme illumination , pose and head motion variation .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 71, "end": 77}, "tail": {"text": "fully automatic recognition system", "start": 14, "end": 48}}], "evaluate for": [{"head": {"text": "video sequences", "start": 139, "end": 154}, "tail": {"text": "fully automatic recognition system", "start": 14, "end": 48}}], "feature of": [{"head": {"text": "illumination", "start": 168, "end": 180}, "tail": {"text": "video sequences", "start": 139, "end": 154}}, {"head": {"text": "pose", "start": 65, "end": 69}, "tail": {"text": "video sequences", "start": 139, "end": 154}}, {"head": {"text": "head motion variation", "start": 192, "end": 213}, "tail": {"text": "video sequences", "start": 139, "end": 154}}], "conjunction": [{"head": {"text": "illumination", "start": 168, "end": 180}, "tail": {"text": "pose", "start": 65, "end": 69}}, {"head": {"text": "pose", "start": 65, "end": 69}, "tail": {"text": "head motion variation", "start": 192, "end": 213}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "On this challenging data set our system consistently demonstrated a nearly perfect recognition rate -LRB- over 99.7 % on all three databases -RRB- , significantly out-performing state-of-the-art commercial software and methods from the literature .", "output": {"relations": {"evaluate for": [{"head": {"text": "data set", "start": 20, "end": 28}, "tail": {"text": "system", "start": 33, "end": 39}}, {"head": {"text": "recognition rate", "start": 83, "end": 99}, "tail": {"text": "system", "start": 33, "end": 39}}], "compare": [{"head": {"text": "system", "start": 33, "end": 39}, "tail": {"text": "commercial software", "start": 195, "end": 214}}, {"head": {"text": "system", "start": 33, "end": 39}, "tail": {"text": "methods", "start": 219, "end": 226}}], "conjunction": [{"head": {"text": "commercial software", "start": 195, "end": 214}, "tail": {"text": "methods", "start": 219, "end": 226}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We present Minimum Bayes-Risk -LRB- MBR -RRB- decoding for statistical machine translation .", "output": {"relations": {"used for": [{"head": {"text": "Minimum Bayes-Risk -LRB- MBR -RRB- decoding", "start": 11, "end": 54}, "tail": {"text": "statistical machine translation", "start": 59, "end": 90}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This statistical approach aims to minimize expected loss of translation errors under loss functions that measure translation performance .", "output": {"relations": {"evaluate for": [{"head": {"text": "loss functions", "start": 85, "end": 99}, "tail": {"text": "translation", "start": 60, "end": 71}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We describe a hierarchy of loss functions that incorporate different levels of linguistic information from word strings , word-to-word alignments from an MT system , and syntactic structure from parse-trees of source and target language sentences .", "output": {"relations": {"used for": [{"head": {"text": "linguistic information", "start": 79, "end": 101}, "tail": {"text": "loss functions", "start": 27, "end": 41}}, {"head": {"text": "word-to-word alignments", "start": 122, "end": 145}, "tail": {"text": "loss functions", "start": 27, "end": 41}}, {"head": {"text": "syntactic structure", "start": 170, "end": 189}, "tail": {"text": "loss functions", "start": 27, "end": 41}}], "part of": [{"head": {"text": "word-to-word alignments", "start": 122, "end": 145}, "tail": {"text": "MT system", "start": 154, "end": 163}}, {"head": {"text": "parse-trees", "start": 195, "end": 206}, "tail": {"text": "syntactic structure", "start": 170, "end": 189}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We report the performance of the MBR decoders on a Chinese-to-English translation task .", "output": {"relations": {"used for": [{"head": {"text": "MBR decoders", "start": 33, "end": 45}, "tail": {"text": "Chinese-to-English translation task", "start": 51, "end": 86}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our results show that MBR decoding can be used to tune statistical MT performance for specific loss functions .", "output": {"relations": {"used for": [{"head": {"text": "MBR decoding", "start": 22, "end": 34}, "tail": {"text": "statistical MT", "start": 55, "end": 69}}, {"head": {"text": "MBR decoding", "start": 22, "end": 34}, "tail": {"text": "loss functions", "start": 95, "end": 109}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper presents a critical discussion of the various approaches that have been used in the evaluation of Natural Language systems .", "output": {"relations": {"used for": [{"head": {"text": "approaches", "start": 57, "end": 67}, "tail": {"text": "evaluation of Natural Language systems", "start": 95, "end": 133}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We conclude that previous approaches have neglected to evaluate systems in the context of their use , e.g. solving a task requiring data retrieval .", "output": {"relations": {"evaluate for": [{"head": {"text": "approaches", "start": 26, "end": 36}, "tail": {"text": "systems", "start": 64, "end": 71}}], "used for": [{"head": {"text": "systems", "start": 64, "end": 71}, "tail": {"text": "task", "start": 117, "end": 121}}], "part of": [{"head": {"text": "data retrieval", "start": 132, "end": 146}, "tail": {"text": "task", "start": 117, "end": 121}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In the second half of the paper , we report a laboratory study using the Wizard of Oz technique to identify NL requirements for carrying out this task .", "output": {"relations": {"used for": [{"head": {"text": "Wizard of Oz technique", "start": 73, "end": 95}, "tail": {"text": "NL requirements", "start": 108, "end": 123}}, {"head": {"text": "Wizard of Oz technique", "start": 73, "end": 95}, "tail": {"text": "task", "start": 146, "end": 150}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We evaluate the demands that task dialogues collected using this technique , place upon a prototype Natural Language system .", "output": {"relations": {"used for": [{"head": {"text": "task dialogues", "start": 29, "end": 43}, "tail": {"text": "prototype Natural Language system", "start": 90, "end": 123}}, {"head": {"text": "technique", "start": 65, "end": 74}, "tail": {"text": "task dialogues", "start": 29, "end": 43}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We present results on addressee identification in four-participants face-to-face meetings using Bayesian Network and Naive Bayes classifiers .", "output": {"relations": {"used for": [{"head": {"text": "Bayesian Network", "start": 96, "end": 112}, "tail": {"text": "addressee identification in four-participants face-to-face meetings", "start": 22, "end": 89}}, {"head": {"text": "Naive Bayes classifiers", "start": 117, "end": 140}, "tail": {"text": "addressee identification in four-participants face-to-face meetings", "start": 22, "end": 89}}], "conjunction": [{"head": {"text": "Naive Bayes classifiers", "start": 117, "end": 140}, "tail": {"text": "Bayesian Network", "start": 96, "end": 112}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "First , we investigate how well the addressee of a dialogue act can be predicted based on gaze , utterance and conversational context features .", "output": {"relations": {"used for": [{"head": {"text": "gaze", "start": 90, "end": 94}, "tail": {"text": "addressee of a dialogue act", "start": 36, "end": 63}}, {"head": {"text": "utterance", "start": 97, "end": 106}, "tail": {"text": "addressee of a dialogue act", "start": 36, "end": 63}}, {"head": {"text": "conversational context features", "start": 111, "end": 142}, "tail": {"text": "addressee of a dialogue act", "start": 36, "end": 63}}], "conjunction": [{"head": {"text": "gaze", "start": 90, "end": 94}, "tail": {"text": "utterance", "start": 97, "end": 106}}, {"head": {"text": "utterance", "start": 97, "end": 106}, "tail": {"text": "conversational context features", "start": 111, "end": 142}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Both classifiers perform the best when conversational context and utterance features are combined with speaker 's gaze information .", "output": {"relations": {"used for": [{"head": {"text": "conversational context", "start": 39, "end": 61}, "tail": {"text": "classifiers", "start": 5, "end": 16}}, {"head": {"text": "utterance features", "start": 66, "end": 84}, "tail": {"text": "classifiers", "start": 5, "end": 16}}, {"head": {"text": "speaker 's gaze information", "start": 103, "end": 130}, "tail": {"text": "classifiers", "start": 5, "end": 16}}], "conjunction": [{"head": {"text": "conversational context", "start": 39, "end": 61}, "tail": {"text": "utterance features", "start": 66, "end": 84}}, {"head": {"text": "speaker 's gaze information", "start": 103, "end": 130}, "tail": {"text": "utterance features", "start": 66, "end": 84}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Towards deep analysis of compositional classes of paraphrases , we have examined a class-oriented framework for collecting paraphrase examples , in which sentential paraphrases are collected for each paraphrase class separately by means of automatic candidate generation and manual judgement .", "output": {"relations": {"used for": [{"head": {"text": "class-oriented framework", "start": 83, "end": 107}, "tail": {"text": "compositional classes of paraphrases", "start": 25, "end": 61}}, {"head": {"text": "class-oriented framework", "start": 83, "end": 107}, "tail": {"text": "paraphrase examples", "start": 123, "end": 142}}, {"head": {"text": "automatic candidate generation", "start": 240, "end": 270}, "tail": {"text": "sentential paraphrases", "start": 154, "end": 176}}, {"head": {"text": "manual judgement", "start": 275, "end": 291}, "tail": {"text": "sentential paraphrases", "start": 154, "end": 176}}], "conjunction": [{"head": {"text": "automatic candidate generation", "start": 240, "end": 270}, "tail": {"text": "manual judgement", "start": 275, "end": 291}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The purpose of this research is to test the efficacy of applying automated evaluation techniques , originally devised for the evaluation of human language learners , to the output of machine translation -LRB- MT -RRB- systems .", "output": {"relations": {"used for": [{"head": {"text": "automated evaluation techniques", "start": 65, "end": 96}, "tail": {"text": "evaluation of human language learners", "start": 126, "end": 163}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We believe that these evaluation techniques will provide information about both the human language learning process , the translation process and the development of machine translation systems .", "output": {"relations": {"used for": [{"head": {"text": "evaluation techniques", "start": 22, "end": 43}, "tail": {"text": "human language learning process", "start": 84, "end": 115}}, {"head": {"text": "evaluation techniques", "start": 22, "end": 43}, "tail": {"text": "translation process", "start": 122, "end": 141}}, {"head": {"text": "evaluation techniques", "start": 22, "end": 43}, "tail": {"text": "machine translation systems", "start": 165, "end": 192}}], "conjunction": [{"head": {"text": "human language learning process", "start": 84, "end": 115}, "tail": {"text": "translation process", "start": 122, "end": 141}}, {"head": {"text": "translation process", "start": 122, "end": 141}, "tail": {"text": "machine translation systems", "start": 165, "end": 192}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A language learning experiment showed that assessors can differentiate native from non-native language essays in less than 100 words .", "output": {"relations": {"evaluate for": [{"head": {"text": "language learning", "start": 2, "end": 19}, "tail": {"text": "assessors", "start": 43, "end": 52}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Some of the extracts were expert human translations , others were machine translation outputs .", "output": {"relations": {"conjunction": [{"head": {"text": "machine translation outputs", "start": 66, "end": 93}, "tail": {"text": "expert human translations", "start": 26, "end": 51}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The subjects were given three minutes per extract to determine whether they believed the sample output to be an expert human translation or a machine translation .", "output": {"relations": {"compare": [{"head": {"text": "expert human translation", "start": 112, "end": 136}, "tail": {"text": "machine translation", "start": 142, "end": 161}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper presents a machine learning approach to bare slice disambiguation in dialogue .", "output": {"relations": {"used for": [{"head": {"text": "machine learning approach", "start": 22, "end": 47}, "tail": {"text": "bare slice disambiguation", "start": 51, "end": 76}}, {"head": {"text": "dialogue", "start": 80, "end": 88}, "tail": {"text": "bare slice disambiguation", "start": 51, "end": 76}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We extract a set of heuristic principles from a corpus-based sample and formulate them as probabilistic Horn clauses .", "output": {"relations": {"used for": [{"head": {"text": "corpus-based sample", "start": 48, "end": 67}, "tail": {"text": "heuristic principles", "start": 20, "end": 40}}], "feature of": [{"head": {"text": "probabilistic Horn clauses", "start": 90, "end": 116}, "tail": {"text": "heuristic principles", "start": 20, "end": 40}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We then use the predicates of such clauses to create a set of domain independent features to annotate an input dataset , and run two different machine learning algorithms : SLIPPER , a rule-based learning algorithm , and TiMBL , a memory-based system .", "output": {"relations": {"hyponym of": [{"head": {"text": "SLIPPER", "start": 173, "end": 180}, "tail": {"text": "rule-based learning algorithm", "start": 185, "end": 214}}, {"head": {"text": "TiMBL", "start": 221, "end": 226}, "tail": {"text": "memory-based system", "start": 231, "end": 250}}], "part of": [{"head": {"text": "rule-based learning algorithm", "start": 185, "end": 214}, "tail": {"text": "machine learning algorithms", "start": 143, "end": 170}}, {"head": {"text": "memory-based system", "start": 231, "end": 250}, "tail": {"text": "machine learning algorithms", "start": 143, "end": 170}}], "compare": [{"head": {"text": "rule-based learning algorithm", "start": 185, "end": 214}, "tail": {"text": "memory-based system", "start": 231, "end": 250}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The results show that the features in terms of which we formulate our heuristic principles have significant predictive power , and that rules that closely resemble our Horn clauses can be learnt automatically from these features .", "output": {"relations": {"feature of": [{"head": {"text": "features", "start": 26, "end": 34}, "tail": {"text": "heuristic principles", "start": 70, "end": 90}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We suggest a new goal and evaluation criterion for word similarity measures .", "output": {"relations": {"used for": [{"head": {"text": "evaluation criterion", "start": 26, "end": 46}, "tail": {"text": "word similarity measures", "start": 51, "end": 75}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The new criterion -- meaning-entailing substitutability -- fits the needs of semantic-oriented NLP applications and can be evaluated directly -LRB- independent of an application -RRB- at a good level of human agreement .", "output": {"relations": {"used for": [{"head": {"text": "meaning-entailing substitutability", "start": 21, "end": 55}, "tail": {"text": "semantic-oriented NLP applications", "start": 77, "end": 111}}], "evaluate for": [{"head": {"text": "human agreement", "start": 203, "end": 218}, "tail": {"text": "meaning-entailing substitutability", "start": 21, "end": 55}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Motivated by this semantic criterion we analyze the empirical quality of distributional word feature vectors and its impact on word similarity results , proposing an objective measure for evaluating feature vector quality .", "output": {"relations": {"evaluate for": [{"head": {"text": "semantic criterion", "start": 18, "end": 36}, "tail": {"text": "distributional word feature vectors", "start": 73, "end": 108}}, {"head": {"text": "measure", "start": 176, "end": 183}, "tail": {"text": "feature vector quality", "start": 199, "end": 221}}], "used for": [{"head": {"text": "distributional word feature vectors", "start": 73, "end": 108}, "tail": {"text": "word similarity", "start": 127, "end": 142}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Finally , a novel feature weighting and selection function is presented , which yields superior feature vectors and better word similarity performance .", "output": {"relations": {"used for": [{"head": {"text": "feature weighting and selection function", "start": 18, "end": 58}, "tail": {"text": "feature vectors", "start": 96, "end": 111}}, {"head": {"text": "feature weighting and selection function", "start": 18, "end": 58}, "tail": {"text": "word similarity", "start": 123, "end": 138}}], "conjunction": [{"head": {"text": "feature vectors", "start": 96, "end": 111}, "tail": {"text": "word similarity", "start": 123, "end": 138}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This phenomenon causes many image processing techniques to fail as they assume the presence of only one layer at each examined site e.g. motion estimation and object recognition .", "output": {"relations": {"conjunction": [{"head": {"text": "motion estimation", "start": 137, "end": 154}, "tail": {"text": "object recognition", "start": 159, "end": 177}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This work presents an automated technique for detecting reflections in image sequences by analyzing motion trajectories of feature points .", "output": {"relations": {"used for": [{"head": {"text": "technique", "start": 32, "end": 41}, "tail": {"text": "detecting reflections in image sequences", "start": 46, "end": 86}}, {"head": {"text": "motion trajectories", "start": 100, "end": 119}, "tail": {"text": "technique", "start": 32, "end": 41}}], "feature of": [{"head": {"text": "feature points", "start": 123, "end": 137}, "tail": {"text": "motion trajectories", "start": 100, "end": 119}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "It models reflection as regions containing two different layers moving over each other .", "output": {"relations": {"used for": [{"head": {"text": "It", "start": 0, "end": 2}, "tail": {"text": "reflection", "start": 10, "end": 20}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We present a strong detector based on combining a set of weak detectors .", "output": {"relations": {"used for": [{"head": {"text": "detectors", "start": 62, "end": 71}, "tail": {"text": "detector", "start": 20, "end": 28}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We use novel priors , generate sparse and dense detection maps and our results show high detection rate with rejection to pathological motion and occlusion .", "output": {"relations": {"used for": [{"head": {"text": "priors", "start": 13, "end": 19}, "tail": {"text": "sparse and dense detection maps", "start": 31, "end": 62}}], "conjunction": [{"head": {"text": "pathological motion", "start": 122, "end": 141}, "tail": {"text": "occlusion", "start": 146, "end": 155}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper considers the problem of reconstructing the motion of a 3D articulated tree from 2D point correspondences subject to some temporal prior .", "output": {"relations": {"used for": [{"head": {"text": "2D point correspondences", "start": 92, "end": 116}, "tail": {"text": "reconstructing the motion of a 3D articulated tree", "start": 36, "end": 86}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Hitherto , smooth motion has been encouraged using a trajectory basis , yielding a hard combinatorial problem with time complexity growing exponentially in the number of frames .", "output": {"relations": {"used for": [{"head": {"text": "trajectory basis", "start": 53, "end": 69}, "tail": {"text": "smooth motion", "start": 11, "end": 24}}], "evaluate for": [{"head": {"text": "time complexity", "start": 115, "end": 130}, "tail": {"text": "hard combinatorial problem", "start": 83, "end": 109}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Branch and bound strategies have previously attempted to curb this complexity whilst maintaining global optimality .", "output": {"relations": {"used for": [{"head": {"text": "Branch and bound strategies", "start": 0, "end": 27}, "tail": {"text": "complexity", "start": 67, "end": 77}}], "feature of": [{"head": {"text": "global optimality", "start": 97, "end": 114}, "tail": {"text": "Branch and bound strategies", "start": 0, "end": 27}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "However , they provide no guarantee of being more efficient than exhaustive search .", "output": {"relations": {"compare": [{"head": {"text": "they", "start": 10, "end": 14}, "tail": {"text": "exhaustive search", "start": 65, "end": 82}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Extension to affine projection enables reconstruction without estimating cameras .", "output": {"relations": {"used for": [{"head": {"text": "affine projection", "start": 13, "end": 30}, "tail": {"text": "reconstruction", "start": 39, "end": 53}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Topical blog post retrieval is the task of ranking blog posts with respect to their relevance for a given topic .", "output": {"relations": {"hyponym of": [{"head": {"text": "Topical blog post retrieval", "start": 0, "end": 27}, "tail": {"text": "ranking blog posts", "start": 43, "end": 61}}], "feature of": [{"head": {"text": "relevance", "start": 84, "end": 93}, "tail": {"text": "blog posts", "start": 51, "end": 61}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "To improve topical blog post retrieval we incorporate textual credibility indicators in the retrieval process .", "output": {"relations": {"used for": [{"head": {"text": "textual credibility indicators", "start": 54, "end": 84}, "tail": {"text": "topical blog post retrieval", "start": 11, "end": 38}}], "part of": [{"head": {"text": "textual credibility indicators", "start": 54, "end": 84}, "tail": {"text": "retrieval process", "start": 92, "end": 109}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We describe how to estimate these indicators and how to integrate them into a retrieval approach based on language models .", "output": {"relations": {"part of": [{"head": {"text": "them", "start": 66, "end": 70}, "tail": {"text": "retrieval approach", "start": 78, "end": 96}}], "used for": [{"head": {"text": "language models", "start": 106, "end": 121}, "tail": {"text": "them", "start": 66, "end": 70}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Experiments on the TREC Blog track test set show that both groups of credibility indicators significantly improve retrieval effectiveness ; the best performance is achieved when combining them .", "output": {"relations": {"evaluate for": [{"head": {"text": "TREC Blog track test set", "start": 19, "end": 43}, "tail": {"text": "credibility indicators", "start": 69, "end": 91}}, {"head": {"text": "retrieval effectiveness", "start": 114, "end": 137}, "tail": {"text": "credibility indicators", "start": 69, "end": 91}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We investigate the problem of learning to predict moves in the board game of Go from game records of expert players .", "output": {"relations": {"used for": [{"head": {"text": "game records of expert players", "start": 85, "end": 115}, "tail": {"text": "board game of Go", "start": 63, "end": 79}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This distribution has numerous applications in computer Go , including serving as an efficient stand-alone Go player .", "output": {"relations": {"used for": [{"head": {"text": "distribution", "start": 5, "end": 17}, "tail": {"text": "computer Go", "start": 47, "end": 58}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "It would also be effective as a move selector and move sorter for game tree search and as a training tool for Go players .", "output": {"relations": {"used for": [{"head": {"text": "It", "start": 0, "end": 2}, "tail": {"text": "move selector", "start": 32, "end": 45}}, {"head": {"text": "It", "start": 0, "end": 2}, "tail": {"text": "move sorter", "start": 50, "end": 61}}, {"head": {"text": "It", "start": 0, "end": 2}, "tail": {"text": "training tool", "start": 92, "end": 105}}, {"head": {"text": "move selector", "start": 32, "end": 45}, "tail": {"text": "game tree search", "start": 66, "end": 82}}, {"head": {"text": "move sorter", "start": 50, "end": 61}, "tail": {"text": "game tree search", "start": 66, "end": 82}}, {"head": {"text": "training tool", "start": 92, "end": 105}, "tail": {"text": "Go players", "start": 110, "end": 120}}], "conjunction": [{"head": {"text": "move selector", "start": 32, "end": 45}, "tail": {"text": "move sorter", "start": 50, "end": 61}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our method has two major components : a -RRB- a pattern extraction scheme for efficiently harvesting patterns of given size and shape from expert game records and b -RRB- a Bayesian learning algorithm -LRB- in two variants -RRB- that learns a distribution over the values of a move given a board position based on the local pattern context .", "output": {"relations": {"part of": [{"head": {"text": "pattern extraction scheme", "start": 48, "end": 73}, "tail": {"text": "method", "start": 4, "end": 10}}, {"head": {"text": "Bayesian learning algorithm", "start": 173, "end": 200}, "tail": {"text": "method", "start": 4, "end": 10}}], "conjunction": [{"head": {"text": "pattern extraction scheme", "start": 48, "end": 73}, "tail": {"text": "Bayesian learning algorithm", "start": 173, "end": 200}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The system is trained on 181,000 expert games and shows excellent prediction performance as indicated by its ability to perfectly predict the moves made by professional Go players in 34 % of test positions .", "output": {"relations": {"used for": [{"head": {"text": "expert games", "start": 33, "end": 45}, "tail": {"text": "system", "start": 4, "end": 10}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We present a novel approach for automatically acquiring English topic signatures .", "output": {"relations": {"used for": [{"head": {"text": "approach", "start": 19, "end": 27}, "tail": {"text": "automatically acquiring English topic signatures", "start": 32, "end": 80}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Topic signatures can be useful in a number of Natural Language Processing -LRB- NLP -RRB- applications , such as Word Sense Disambiguation -LRB- WSD -RRB- and Text Summarisation .", "output": {"relations": {"used for": [{"head": {"text": "Topic signatures", "start": 0, "end": 16}, "tail": {"text": "Natural Language Processing -LRB- NLP -RRB- applications", "start": 46, "end": 102}}, {"head": {"text": "Topic signatures", "start": 0, "end": 16}, "tail": {"text": "Word Sense Disambiguation -LRB- WSD -RRB-", "start": 113, "end": 154}}, {"head": {"text": "Topic signatures", "start": 0, "end": 16}, "tail": {"text": "Text Summarisation", "start": 159, "end": 177}}], "hyponym of": [{"head": {"text": "Word Sense Disambiguation -LRB- WSD -RRB-", "start": 113, "end": 154}, "tail": {"text": "Natural Language Processing -LRB- NLP -RRB- applications", "start": 46, "end": 102}}, {"head": {"text": "Text Summarisation", "start": 159, "end": 177}, "tail": {"text": "Natural Language Processing -LRB- NLP -RRB- applications", "start": 46, "end": 102}}], "conjunction": [{"head": {"text": "Word Sense Disambiguation -LRB- WSD -RRB-", "start": 113, "end": 154}, "tail": {"text": "Text Summarisation", "start": 159, "end": 177}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our method takes advantage of the different way in which word senses are lexicalised in English and Chinese , and also exploits the large amount of Chinese text available in corpora and on the Web .", "output": {"relations": {"part of": [{"head": {"text": "Chinese text", "start": 148, "end": 160}, "tail": {"text": "corpora", "start": 174, "end": 181}}, {"head": {"text": "Chinese text", "start": 148, "end": 160}, "tail": {"text": "Web", "start": 193, "end": 196}}], "conjunction": [{"head": {"text": "corpora", "start": 174, "end": 181}, "tail": {"text": "Web", "start": 193, "end": 196}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We evaluated the topic signatures on a WSD task , where we trained a second-order vector cooccurrence algorithm on standard WSD datasets , with promising results .", "output": {"relations": {"evaluate for": [{"head": {"text": "WSD task", "start": 39, "end": 47}, "tail": {"text": "topic signatures", "start": 17, "end": 33}}], "used for": [{"head": {"text": "WSD datasets", "start": 124, "end": 136}, "tail": {"text": "second-order vector cooccurrence algorithm", "start": 69, "end": 111}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Joint matrix triangularization is often used for estimating the joint eigenstructure of a set M of matrices , with applications in signal processing and machine learning .", "output": {"relations": {"used for": [{"head": {"text": "Joint matrix triangularization", "start": 0, "end": 30}, "tail": {"text": "joint eigenstructure", "start": 64, "end": 84}}, {"head": {"text": "joint eigenstructure", "start": 64, "end": 84}, "tail": {"text": "signal processing", "start": 131, "end": 148}}, {"head": {"text": "joint eigenstructure", "start": 64, "end": 84}, "tail": {"text": "machine learning", "start": 153, "end": 169}}], "conjunction": [{"head": {"text": "signal processing", "start": 131, "end": 148}, "tail": {"text": "machine learning", "start": 153, "end": 169}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our main result is a first-order upper bound on the distance between any approximate joint triangularizer of the matrices in M ' and any exact joint triangularizer of the matrices in M .", "output": {"relations": {"conjunction": [{"head": {"text": "approximate joint triangularizer", "start": 73, "end": 105}, "tail": {"text": "exact joint triangularizer", "start": 137, "end": 163}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "To our knowledge , this is the first a posteriori bound for joint matrix decomposition .", "output": {"relations": {"used for": [{"head": {"text": "posteriori bound", "start": 39, "end": 55}, "tail": {"text": "joint matrix decomposition", "start": 60, "end": 86}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The psycholinguistic literature provides evidence for syntactic priming , i.e. , the tendency to repeat structures .", "output": {"relations": {"used for": [{"head": {"text": "psycholinguistic literature", "start": 4, "end": 31}, "tail": {"text": "syntactic priming", "start": 54, "end": 71}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper describes a method for incorporating priming into an incremental probabilistic parser .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 23, "end": 29}, "tail": {"text": "priming", "start": 48, "end": 55}}, {"head": {"text": "priming", "start": 48, "end": 55}, "tail": {"text": "incremental probabilistic parser", "start": 64, "end": 96}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "These models simulate the reading time advantage for parallel structures found in human data , and also yield a small increase in overall parsing accuracy .", "output": {"relations": {"part of": [{"head": {"text": "parallel structures", "start": 53, "end": 72}, "tail": {"text": "human data", "start": 82, "end": 92}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Learned confidence measures gain increasing importance for outlier removal and quality improvement in stereo vision .", "output": {"relations": {"used for": [{"head": {"text": "Learned confidence measures", "start": 0, "end": 27}, "tail": {"text": "outlier removal", "start": 59, "end": 74}}, {"head": {"text": "Learned confidence measures", "start": 0, "end": 27}, "tail": {"text": "quality improvement", "start": 79, "end": 98}}], "conjunction": [{"head": {"text": "outlier removal", "start": 59, "end": 74}, "tail": {"text": "quality improvement", "start": 79, "end": 98}}], "part of": [{"head": {"text": "outlier removal", "start": 59, "end": 74}, "tail": {"text": "stereo vision", "start": 102, "end": 115}}, {"head": {"text": "quality improvement", "start": 79, "end": 98}, "tail": {"text": "stereo vision", "start": 102, "end": 115}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "However , acquiring the necessary training data is typically a tedious and time consuming task that involves manual interaction , active sensing devices and/or synthetic scenes .", "output": {"relations": {"used for": [{"head": {"text": "manual interaction", "start": 109, "end": 127}, "tail": {"text": "task", "start": 90, "end": 94}}, {"head": {"text": "active sensing devices", "start": 130, "end": 152}, "tail": {"text": "task", "start": 90, "end": 94}}, {"head": {"text": "synthetic scenes", "start": 160, "end": 176}, "tail": {"text": "task", "start": 90, "end": 94}}], "conjunction": [{"head": {"text": "manual interaction", "start": 109, "end": 127}, "tail": {"text": "active sensing devices", "start": 130, "end": 152}}, {"head": {"text": "active sensing devices", "start": 130, "end": 152}, "tail": {"text": "synthetic scenes", "start": 160, "end": 176}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The key idea of our approach is to use different view points for reasoning about contradictions and consistencies between multiple depth maps generated with the same stereo algorithm .", "output": {"relations": {"used for": [{"head": {"text": "view points", "start": 49, "end": 60}, "tail": {"text": "approach", "start": 20, "end": 28}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Among other experiments , we demonstrate the potential of our approach by boosting the performance of three learned confidence measures on the KITTI2012 dataset by simply training them on a vast amount of automatically generated training data rather than a limited amount of laser ground truth data .", "output": {"relations": {"used for": [{"head": {"text": "approach", "start": 62, "end": 70}, "tail": {"text": "learned confidence measures", "start": 108, "end": 135}}, {"head": {"text": "automatically generated training data", "start": 205, "end": 242}, "tail": {"text": "them", "start": 180, "end": 184}}], "evaluate for": [{"head": {"text": "KITTI2012 dataset", "start": 143, "end": 160}, "tail": {"text": "learned confidence measures", "start": 108, "end": 135}}], "compare": [{"head": {"text": "laser ground truth data", "start": 275, "end": 298}, "tail": {"text": "automatically generated training data", "start": 205, "end": 242}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "An important area of learning in autonomous agents is the ability to learn domain-speciic models of actions to be used by planning systems .", "output": {"relations": {"used for": [{"head": {"text": "learning in autonomous agents", "start": 21, "end": 50}, "tail": {"text": "domain-speciic models of actions", "start": 75, "end": 107}}, {"head": {"text": "planning systems", "start": 122, "end": 138}, "tail": {"text": "domain-speciic models of actions", "start": 75, "end": 107}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "These methods diier from previous work in the area in two ways : the use of an action model formalism which is better suited to the needs of a re-active agent , and successful implementation of noise-handling mechanisms .", "output": {"relations": {"used for": [{"head": {"text": "action model formalism", "start": 79, "end": 101}, "tail": {"text": "methods", "start": 6, "end": 13}}, {"head": {"text": "action model formalism", "start": 79, "end": 101}, "tail": {"text": "re-active agent", "start": 143, "end": 158}}, {"head": {"text": "noise-handling mechanisms", "start": 194, "end": 219}, "tail": {"text": "methods", "start": 6, "end": 13}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Training instances are generated from experience and observation , and a variant of GOLEM is used to learn action models from these instances .", "output": {"relations": {"used for": [{"head": {"text": "GOLEM", "start": 84, "end": 89}, "tail": {"text": "action models", "start": 107, "end": 120}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The integrated learning system has been experimentally validated in simulated construction and ooce domains .", "output": {"relations": {"evaluate for": [{"head": {"text": "simulated construction", "start": 68, "end": 90}, "tail": {"text": "integrated learning system", "start": 4, "end": 30}}, {"head": {"text": "ooce domains", "start": 95, "end": 107}, "tail": {"text": "integrated learning system", "start": 4, "end": 30}}], "conjunction": [{"head": {"text": "simulated construction", "start": 68, "end": 90}, "tail": {"text": "ooce domains", "start": 95, "end": 107}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper describes FERRET , an interactive question-answering -LRB- Q/A -RRB- system designed to address the challenges of integrating automatic Q/A applications into real-world environments .", "output": {"relations": {"hyponym of": [{"head": {"text": "FERRET", "start": 21, "end": 27}, "tail": {"text": "interactive question-answering -LRB- Q/A -RRB- system", "start": 33, "end": 86}}], "used for": [{"head": {"text": "FERRET", "start": 21, "end": 27}, "tail": {"text": "integrating automatic Q/A applications into real-world environments", "start": 125, "end": 192}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "FERRET utilizes a novel approach to Q/A known as predictive questioning which attempts to identify the questions -LRB- and answers -RRB- that users need by analyzing how a user interacts with a system while gathering information related to a particular scenario .", "output": {"relations": {"used for": [{"head": {"text": "approach", "start": 24, "end": 32}, "tail": {"text": "FERRET", "start": 0, "end": 6}}, {"head": {"text": "approach", "start": 24, "end": 32}, "tail": {"text": "Q/A", "start": 36, "end": 39}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In order to build robust automatic abstracting systems , there is a need for better training resources than are currently available .", "output": {"relations": {"used for": [{"head": {"text": "training resources", "start": 84, "end": 102}, "tail": {"text": "automatic abstracting systems", "start": 25, "end": 54}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , we introduce an annotation scheme for scientific articles which can be used to build such a resource in a consistent way .", "output": {"relations": {"used for": [{"head": {"text": "annotation scheme", "start": 32, "end": 49}, "tail": {"text": "scientific articles", "start": 54, "end": 73}}, {"head": {"text": "annotation scheme", "start": 32, "end": 49}, "tail": {"text": "resource", "start": 108, "end": 116}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The seven categories of the scheme are based on rhetorical moves of argumentation .", "output": {"relations": {"used for": [{"head": {"text": "rhetorical moves of argumentation", "start": 48, "end": 81}, "tail": {"text": "scheme", "start": 28, "end": 34}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The automated segmentation of images into semantically meaningful parts requires shape information since low-level feature analysis alone often fails to reach this goal .", "output": {"relations": {"used for": [{"head": {"text": "images", "start": 30, "end": 36}, "tail": {"text": "automated segmentation", "start": 4, "end": 26}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We introduce a novel method of shape constrained image segmentation which is based on mixtures of feature distributions for color and texture as well as probabilistic shape knowledge .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 21, "end": 27}, "tail": {"text": "shape constrained image segmentation", "start": 31, "end": 67}}, {"head": {"text": "mixtures of feature distributions", "start": 86, "end": 119}, "tail": {"text": "method", "start": 21, "end": 27}}, {"head": {"text": "mixtures of feature distributions", "start": 86, "end": 119}, "tail": {"text": "color", "start": 124, "end": 129}}, {"head": {"text": "mixtures of feature distributions", "start": 86, "end": 119}, "tail": {"text": "texture", "start": 134, "end": 141}}, {"head": {"text": "mixtures of feature distributions", "start": 86, "end": 119}, "tail": {"text": "probabilistic shape knowledge", "start": 153, "end": 182}}], "conjunction": [{"head": {"text": "color", "start": 124, "end": 129}, "tail": {"text": "texture", "start": 134, "end": 141}}, {"head": {"text": "texture", "start": 134, "end": 141}, "tail": {"text": "probabilistic shape knowledge", "start": 153, "end": 182}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The combined approach is formulated in the framework of Bayesian statistics to account for the robust-ness requirement in image understanding .", "output": {"relations": {"used for": [{"head": {"text": "approach", "start": 13, "end": 21}, "tail": {"text": "robust-ness requirement in image understanding", "start": 95, "end": 141}}, {"head": {"text": "Bayesian statistics", "start": 56, "end": 75}, "tail": {"text": "approach", "start": 13, "end": 21}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The goal of this work is the enrichment of human-machine interactions in a natural language environment .", "output": {"relations": {"feature of": [{"head": {"text": "natural language environment", "start": 75, "end": 103}, "tail": {"text": "human-machine interactions", "start": 43, "end": 69}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper highlights a particular class of miscommunication -- reference problems -- by describing a case study and techniques for avoiding failures of reference .", "output": {"relations": {"hyponym of": [{"head": {"text": "reference problems", "start": 64, "end": 82}, "tail": {"text": "miscommunication", "start": 44, "end": 60}}], "used for": [{"head": {"text": "techniques", "start": 117, "end": 127}, "tail": {"text": "failures of reference", "start": 141, "end": 162}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper examines the benefits of system combination for unsupervised WSD .", "output": {"relations": {"used for": [{"head": {"text": "system combination", "start": 36, "end": 54}, "tail": {"text": "unsupervised WSD", "start": 59, "end": 75}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We investigate several voting - and arbiter-based combination strategies over a diverse pool of unsupervised WSD systems .", "output": {"relations": {"used for": [{"head": {"text": "voting - and arbiter-based combination strategies", "start": 23, "end": 72}, "tail": {"text": "unsupervised WSD systems", "start": 96, "end": 120}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our combination methods rely on predominant senses which are derived automatically from raw text .", "output": {"relations": {"used for": [{"head": {"text": "predominant senses", "start": 32, "end": 50}, "tail": {"text": "combination methods", "start": 4, "end": 23}}, {"head": {"text": "raw text", "start": 88, "end": 96}, "tail": {"text": "predominant senses", "start": 32, "end": 50}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Experiments using the SemCor and Senseval-3 data sets demonstrate that our ensembles yield significantly better results when compared with state-of-the-art .", "output": {"relations": {"evaluate for": [{"head": {"text": "SemCor and Senseval-3 data sets", "start": 22, "end": 53}, "tail": {"text": "ensembles", "start": 75, "end": 84}}, {"head": {"text": "SemCor and Senseval-3 data sets", "start": 22, "end": 53}, "tail": {"text": "state-of-the-art", "start": 139, "end": 155}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The applicability of many current information extraction techniques is severely limited by the need for supervised training data .", "output": {"relations": {"used for": [{"head": {"text": "supervised training data", "start": 104, "end": 128}, "tail": {"text": "information extraction techniques", "start": 34, "end": 67}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We demonstrate that for certain field structured extraction tasks , such as classified advertisements and bibliographic citations , small amounts of prior knowledge can be used to learn effective models in a primarily unsupervised fashion .", "output": {"relations": {"hyponym of": [{"head": {"text": "classified advertisements", "start": 76, "end": 101}, "tail": {"text": "field structured extraction tasks", "start": 32, "end": 65}}, {"head": {"text": "bibliographic citations", "start": 106, "end": 129}, "tail": {"text": "field structured extraction tasks", "start": 32, "end": 65}}], "conjunction": [{"head": {"text": "classified advertisements", "start": 76, "end": 101}, "tail": {"text": "bibliographic citations", "start": 106, "end": 129}}], "used for": [{"head": {"text": "prior knowledge", "start": 149, "end": 164}, "tail": {"text": "field structured extraction tasks", "start": 32, "end": 65}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Although hidden Markov models -LRB- HMMs -RRB- provide a suitable generative model for field structured text , general unsupervised HMM learning fails to learn useful structure in either of our domains .", "output": {"relations": {"used for": [{"head": {"text": "hidden Markov models -LRB- HMMs -RRB-", "start": 9, "end": 46}, "tail": {"text": "generative model", "start": 66, "end": 82}}, {"head": {"text": "generative model", "start": 66, "end": 82}, "tail": {"text": "field structured text", "start": 87, "end": 108}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In both domains , we found that unsupervised methods can attain accuracies with 400 unlabeled examples comparable to those attained by supervised methods on 50 labeled examples , and that semi-supervised methods can make good use of small amounts of labeled data .", "output": {"relations": {"compare": [{"head": {"text": "unsupervised methods", "start": 32, "end": 52}, "tail": {"text": "supervised methods", "start": 34, "end": 52}}], "evaluate for": [{"head": {"text": "accuracies", "start": 64, "end": 74}, "tail": {"text": "unsupervised methods", "start": 32, "end": 52}}, {"head": {"text": "accuracies", "start": 64, "end": 74}, "tail": {"text": "supervised methods", "start": 34, "end": 52}}], "used for": [{"head": {"text": "unlabeled examples", "start": 84, "end": 102}, "tail": {"text": "unsupervised methods", "start": 32, "end": 52}}, {"head": {"text": "labeled examples", "start": 86, "end": 102}, "tail": {"text": "supervised methods", "start": 34, "end": 52}}, {"head": {"text": "labeled data", "start": 250, "end": 262}, "tail": {"text": "semi-supervised methods", "start": 188, "end": 211}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper gives an overall account of a prototype natural language question answering system , called Chat-80 .", "output": {"relations": {"hyponym of": [{"head": {"text": "Chat-80", "start": 103, "end": 110}, "tail": {"text": "natural language question answering system", "start": 51, "end": 93}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The system is implemented entirely in Prolog , a programming language based on logic .", "output": {"relations": {"used for": [{"head": {"text": "Prolog", "start": 38, "end": 44}, "tail": {"text": "system", "start": 4, "end": 10}}, {"head": {"text": "logic", "start": 79, "end": 84}, "tail": {"text": "programming language", "start": 49, "end": 69}}], "hyponym of": [{"head": {"text": "Prolog", "start": 38, "end": 44}, "tail": {"text": "programming language", "start": 49, "end": 69}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "With the aid of a logic-based grammar formalism called extraposition grammars , Chat-80 translates English questions into the Prolog subset of logic .", "output": {"relations": {"hyponym of": [{"head": {"text": "extraposition grammars", "start": 55, "end": 77}, "tail": {"text": "logic-based grammar formalism", "start": 18, "end": 47}}], "used for": [{"head": {"text": "extraposition grammars", "start": 55, "end": 77}, "tail": {"text": "Chat-80", "start": 80, "end": 87}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The resulting logical expression is then transformed by a planning algorithm into efficient Prolog , cf. query optimisation in a relational database .", "output": {"relations": {"used for": [{"head": {"text": "planning algorithm", "start": 58, "end": 76}, "tail": {"text": "logical expression", "start": 14, "end": 32}}, {"head": {"text": "relational database", "start": 129, "end": 148}, "tail": {"text": "query optimisation", "start": 105, "end": 123}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Human action recognition from well-segmented 3D skeleton data has been intensively studied and attracting an increasing attention .", "output": {"relations": {"used for": [{"head": {"text": "well-segmented 3D skeleton data", "start": 30, "end": 61}, "tail": {"text": "Human action recognition", "start": 0, "end": 24}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Online action detection goes one step further and is more challenging , which identifies the action type and localizes the action positions on the fly from the untrimmed stream .", "output": {"relations": {"used for": [{"head": {"text": "Online action detection", "start": 0, "end": 23}, "tail": {"text": "action type", "start": 93, "end": 104}}, {"head": {"text": "Online action detection", "start": 0, "end": 23}, "tail": {"text": "action positions", "start": 123, "end": 139}}, {"head": {"text": "untrimmed stream", "start": 160, "end": 176}, "tail": {"text": "Online action detection", "start": 0, "end": 23}}], "conjunction": [{"head": {"text": "action type", "start": 93, "end": 104}, "tail": {"text": "action positions", "start": 123, "end": 139}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , we study the problem of online action detection from the streaming skeleton data .", "output": {"relations": {"used for": [{"head": {"text": "streaming skeleton data", "start": 73, "end": 96}, "tail": {"text": "online action detection", "start": 40, "end": 63}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We propose a multi-task end-to-end Joint Classification-Regression Recurrent Neural Network to better explore the action type and temporal localiza-tion information .", "output": {"relations": {"used for": [{"head": {"text": "multi-task end-to-end Joint Classification-Regression Recurrent Neural Network", "start": 13, "end": 91}, "tail": {"text": "action type", "start": 114, "end": 125}}, {"head": {"text": "multi-task end-to-end Joint Classification-Regression Recurrent Neural Network", "start": 13, "end": 91}, "tail": {"text": "temporal localiza-tion information", "start": 130, "end": 164}}], "conjunction": [{"head": {"text": "action type", "start": 114, "end": 125}, "tail": {"text": "temporal localiza-tion information", "start": 130, "end": 164}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "By employing a joint classification and regression optimization objective , this network is capable of automatically localizing the start and end points of actions more accurately .", "output": {"relations": {"used for": [{"head": {"text": "joint classification and regression optimization objective", "start": 15, "end": 73}, "tail": {"text": "network", "start": 81, "end": 88}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Specifically , by leveraging the merits of the deep Long Short-Term Memory -LRB- LSTM -RRB- subnetwork , the proposed model automatically captures the complex long-range temporal dynamics , which naturally avoids the typical sliding window design and thus ensures high computational efficiency .", "output": {"relations": {"used for": [{"head": {"text": "deep Long Short-Term Memory -LRB- LSTM -RRB- subnetwork", "start": 47, "end": 102}, "tail": {"text": "model", "start": 118, "end": 123}}], "feature of": [{"head": {"text": "long-range temporal dynamics", "start": 159, "end": 187}, "tail": {"text": "model", "start": 118, "end": 123}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "To evaluate our proposed model , we build a large streaming video dataset with annotations .", "output": {"relations": {"evaluate for": [{"head": {"text": "streaming video dataset", "start": 50, "end": 73}, "tail": {"text": "model", "start": 25, "end": 30}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Experimental results on our dataset and the public G3D dataset both demonstrate very promising performance of our scheme .", "output": {"relations": {"conjunction": [{"head": {"text": "dataset", "start": 28, "end": 35}, "tail": {"text": "G3D dataset", "start": 51, "end": 62}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The task of machine translation -LRB- MT -RRB- evaluation is closely related to the task of sentence-level semantic equivalence classification .", "output": {"relations": {"conjunction": [{"head": {"text": "machine translation -LRB- MT -RRB- evaluation", "start": 12, "end": 57}, "tail": {"text": "sentence-level semantic equivalence classification", "start": 92, "end": 142}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper investigates the utility of applying standard MT evaluation methods -LRB- BLEU , NIST , WER and PER -RRB- to building classifiers to predict semantic equivalence and entailment .", "output": {"relations": {"used for": [{"head": {"text": "MT evaluation methods", "start": 57, "end": 78}, "tail": {"text": "classifiers", "start": 129, "end": 140}}, {"head": {"text": "classifiers", "start": 129, "end": 140}, "tail": {"text": "semantic equivalence", "start": 152, "end": 172}}, {"head": {"text": "classifiers", "start": 129, "end": 140}, "tail": {"text": "entailment", "start": 177, "end": 187}}], "hyponym of": [{"head": {"text": "BLEU", "start": 85, "end": 89}, "tail": {"text": "MT evaluation methods", "start": 57, "end": 78}}, {"head": {"text": "NIST", "start": 92, "end": 96}, "tail": {"text": "MT evaluation methods", "start": 57, "end": 78}}, {"head": {"text": "WER", "start": 99, "end": 102}, "tail": {"text": "MT evaluation methods", "start": 57, "end": 78}}, {"head": {"text": "PER", "start": 107, "end": 110}, "tail": {"text": "MT evaluation methods", "start": 57, "end": 78}}], "conjunction": [{"head": {"text": "BLEU", "start": 85, "end": 89}, "tail": {"text": "NIST", "start": 92, "end": 96}}, {"head": {"text": "NIST", "start": 92, "end": 96}, "tail": {"text": "WER", "start": 99, "end": 102}}, {"head": {"text": "WER", "start": 99, "end": 102}, "tail": {"text": "PER", "start": 107, "end": 110}}, {"head": {"text": "semantic equivalence", "start": 152, "end": 172}, "tail": {"text": "entailment", "start": 177, "end": 187}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We also introduce a novel classification method based on PER which leverages part of speech information of the words contributing to the word matches and non-matches in the sentence .", "output": {"relations": {"used for": [{"head": {"text": "PER", "start": 57, "end": 60}, "tail": {"text": "classification method", "start": 26, "end": 47}}, {"head": {"text": "PER", "start": 57, "end": 60}, "tail": {"text": "part of speech information", "start": 77, "end": 103}}, {"head": {"text": "part of speech information", "start": 77, "end": 103}, "tail": {"text": "word matches and non-matches", "start": 137, "end": 165}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our results show that MT evaluation techniques are able to produce useful features for paraphrase classification and to a lesser extent entailment .", "output": {"relations": {"used for": [{"head": {"text": "MT evaluation techniques", "start": 22, "end": 46}, "tail": {"text": "features", "start": 74, "end": 82}}, {"head": {"text": "MT evaluation techniques", "start": 22, "end": 46}, "tail": {"text": "paraphrase classification", "start": 87, "end": 112}}, {"head": {"text": "MT evaluation techniques", "start": 22, "end": 46}, "tail": {"text": "entailment", "start": 136, "end": 146}}], "conjunction": [{"head": {"text": "paraphrase classification", "start": 87, "end": 112}, "tail": {"text": "entailment", "start": 136, "end": 146}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our technique gives a substantial improvement in paraphrase classification accuracy over all of the other models used in the experiments .", "output": {"relations": {"compare": [{"head": {"text": "technique", "start": 4, "end": 13}, "tail": {"text": "models", "start": 106, "end": 112}}], "evaluate for": [{"head": {"text": "paraphrase classification accuracy", "start": 49, "end": 83}, "tail": {"text": "technique", "start": 4, "end": 13}}, {"head": {"text": "paraphrase classification accuracy", "start": 49, "end": 83}, "tail": {"text": "models", "start": 106, "end": 112}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Given an object model and a black-box measure of similarity between the model and candidate targets , we consider visual object tracking as a numerical optimization problem .", "output": {"relations": {"used for": [{"head": {"text": "numerical optimization problem", "start": 142, "end": 172}, "tail": {"text": "visual object tracking", "start": 114, "end": 136}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "During normal tracking conditions when the object is visible from frame to frame , local optimization is used to track the local mode of the similarity measure in a parameter space of translation , rotation and scale .", "output": {"relations": {"used for": [{"head": {"text": "local optimization", "start": 83, "end": 101}, "tail": {"text": "local mode of the similarity measure", "start": 123, "end": 159}}, {"head": {"text": "parameter space of translation , rotation and scale", "start": 165, "end": 216}, "tail": {"text": "local mode of the similarity measure", "start": 123, "end": 159}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "However , when the object becomes partially or totally occluded , such local tracking is prone to failure , especially when common prediction techniques like the Kalman filter do not provide a good estimate of object parameters in future frames .", "output": {"relations": {"part of": [{"head": {"text": "Kalman filter", "start": 162, "end": 175}, "tail": {"text": "prediction techniques", "start": 131, "end": 152}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "To recover from these inevitable tracking failures , we consider object detection as a global optimization problem and solve it via Adaptive Simulated Annealing -LRB- ASA -RRB- , a method that avoids becoming trapped at local modes and is much faster than exhaustive search .", "output": {"relations": {"used for": [{"head": {"text": "global optimization problem", "start": 87, "end": 114}, "tail": {"text": "object detection", "start": 65, "end": 81}}, {"head": {"text": "Adaptive Simulated Annealing -LRB- ASA -RRB-", "start": 132, "end": 176}, "tail": {"text": "it", "start": 26, "end": 28}}], "compare": [{"head": {"text": "method", "start": 181, "end": 187}, "tail": {"text": "exhaustive search", "start": 256, "end": 273}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "As a Monte Carlo approach , ASA stochastically samples the parameter space , in contrast to local deterministic search .", "output": {"relations": {"hyponym of": [{"head": {"text": "ASA", "start": 28, "end": 31}, "tail": {"text": "Monte Carlo approach", "start": 5, "end": 25}}], "compare": [{"head": {"text": "ASA", "start": 28, "end": 31}, "tail": {"text": "local deterministic search", "start": 92, "end": 118}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We apply cluster analysis on the sampled parameter space to redetect the object and renew the local tracker .", "output": {"relations": {"used for": [{"head": {"text": "cluster analysis", "start": 9, "end": 25}, "tail": {"text": "sampled parameter space", "start": 33, "end": 56}}, {"head": {"text": "cluster analysis", "start": 9, "end": 25}, "tail": {"text": "local tracker", "start": 94, "end": 107}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our numerical hybrid local and global mode-seeking tracker is validated on challenging airborne videos with heavy occlusion and large camera motions .", "output": {"relations": {"evaluate for": [{"head": {"text": "airborne videos", "start": 87, "end": 102}, "tail": {"text": "numerical hybrid local and global mode-seeking tracker", "start": 4, "end": 58}}], "feature of": [{"head": {"text": "heavy occlusion", "start": 108, "end": 123}, "tail": {"text": "airborne videos", "start": 87, "end": 102}}, {"head": {"text": "camera motions", "start": 134, "end": 148}, "tail": {"text": "airborne videos", "start": 87, "end": 102}}], "conjunction": [{"head": {"text": "heavy occlusion", "start": 108, "end": 123}, "tail": {"text": "camera motions", "start": 134, "end": 148}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our approach outperforms state-of-the-art trackers on the VIVID benchmark datasets .", "output": {"relations": {"compare": [{"head": {"text": "state-of-the-art trackers", "start": 25, "end": 50}, "tail": {"text": "approach", "start": 4, "end": 12}}], "evaluate for": [{"head": {"text": "VIVID benchmark datasets", "start": 58, "end": 82}, "tail": {"text": "approach", "start": 4, "end": 12}}, {"head": {"text": "VIVID benchmark datasets", "start": 58, "end": 82}, "tail": {"text": "state-of-the-art trackers", "start": 25, "end": 50}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Techniques for automatically training modules of a natural language generator have recently been proposed , but a fundamental concern is whether the quality of utterances produced with trainable components can compete with hand-crafted template-based or rule-based approaches .", "output": {"relations": {"used for": [{"head": {"text": "Techniques", "start": 0, "end": 10}, "tail": {"text": "automatically training modules", "start": 15, "end": 45}}], "part of": [{"head": {"text": "automatically training modules", "start": 15, "end": 45}, "tail": {"text": "natural language generator", "start": 51, "end": 77}}], "evaluate for": [{"head": {"text": "utterances", "start": 160, "end": 170}, "tail": {"text": "trainable components", "start": 185, "end": 205}}, {"head": {"text": "utterances", "start": 160, "end": 170}, "tail": {"text": "hand-crafted template-based or rule-based approaches", "start": 223, "end": 275}}], "compare": [{"head": {"text": "trainable components", "start": 185, "end": 205}, "tail": {"text": "hand-crafted template-based or rule-based approaches", "start": 223, "end": 275}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper We experimentally evaluate a trainable sentence planner for a spoken dialogue system by eliciting subjective human judgments .", "output": {"relations": {"used for": [{"head": {"text": "trainable sentence planner", "start": 43, "end": 69}, "tail": {"text": "spoken dialogue system", "start": 76, "end": 98}}], "evaluate for": [{"head": {"text": "subjective human judgments", "start": 112, "end": 138}, "tail": {"text": "trainable sentence planner", "start": 43, "end": 69}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In order to perform an exhaustive comparison , we also evaluate a hand-crafted template-based generation component , two rule-based sentence planners , and two baseline sentence planners .", "output": {"relations": {"conjunction": [{"head": {"text": "hand-crafted template-based generation component", "start": 66, "end": 114}, "tail": {"text": "rule-based sentence planners", "start": 121, "end": 149}}, {"head": {"text": "rule-based sentence planners", "start": 121, "end": 149}, "tail": {"text": "baseline sentence planners", "start": 160, "end": 186}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We show that the trainable sentence planner performs better than the rule-based systems and the baselines , and as well as the hand-crafted system .", "output": {"relations": {"compare": [{"head": {"text": "trainable sentence planner", "start": 17, "end": 43}, "tail": {"text": "rule-based systems", "start": 69, "end": 87}}, {"head": {"text": "trainable sentence planner", "start": 17, "end": 43}, "tail": {"text": "baselines", "start": 96, "end": 105}}, {"head": {"text": "trainable sentence planner", "start": 17, "end": 43}, "tail": {"text": "hand-crafted system", "start": 127, "end": 146}}], "conjunction": [{"head": {"text": "rule-based systems", "start": 69, "end": 87}, "tail": {"text": "baselines", "start": 96, "end": 105}}, {"head": {"text": "baselines", "start": 96, "end": 105}, "tail": {"text": "hand-crafted system", "start": 127, "end": 146}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A new algorithm is proposed for novel view generation in one-to-one teleconferencing applications .", "output": {"relations": {"used for": [{"head": {"text": "algorithm", "start": 6, "end": 15}, "tail": {"text": "novel view generation", "start": 32, "end": 53}}, {"head": {"text": "novel view generation", "start": 32, "end": 53}, "tail": {"text": "one-to-one teleconferencing applications", "start": 57, "end": 97}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Given the video streams acquired by two cameras placed on either side of a computer monitor , the proposed algorithm synthesises images from a virtual camera in arbitrary position -LRB- typically located within the monitor -RRB- to facilitate eye contact .", "output": {"relations": {"used for": [{"head": {"text": "cameras", "start": 40, "end": 47}, "tail": {"text": "video streams", "start": 10, "end": 23}}, {"head": {"text": "algorithm", "start": 107, "end": 116}, "tail": {"text": "eye contact", "start": 243, "end": 254}}, {"head": {"text": "virtual camera", "start": 143, "end": 157}, "tail": {"text": "images", "start": 129, "end": 135}}], "feature of": [{"head": {"text": "arbitrary position", "start": 161, "end": 179}, "tail": {"text": "virtual camera", "start": 143, "end": 157}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our technique is based on an improved , dynamic-programming , stereo algorithm for efficient novel-view generation .", "output": {"relations": {"used for": [{"head": {"text": "technique", "start": 4, "end": 13}, "tail": {"text": "novel-view generation", "start": 93, "end": 114}}, {"head": {"text": "dynamic-programming , stereo algorithm", "start": 40, "end": 78}, "tail": {"text": "technique", "start": 4, "end": 13}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The two main contributions of this paper are : i -RRB- a new type of three-plane graph for dense-stereo dynamic-programming , that encourages correct occlusion labeling ; ii -RRB- a compact geometric derivation for novel-view synthesis by direct projection of the minimum-cost surface .", "output": {"relations": {"used for": [{"head": {"text": "three-plane graph", "start": 69, "end": 86}, "tail": {"text": "dense-stereo dynamic-programming", "start": 91, "end": 123}}, {"head": {"text": "dense-stereo dynamic-programming", "start": 91, "end": 123}, "tail": {"text": "occlusion labeling", "start": 150, "end": 168}}, {"head": {"text": "compact geometric derivation", "start": 182, "end": 210}, "tail": {"text": "novel-view synthesis", "start": 215, "end": 235}}, {"head": {"text": "direct projection of the minimum-cost surface", "start": 239, "end": 284}, "tail": {"text": "compact geometric derivation", "start": 182, "end": 210}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Furthermore , this paper presents a novel algorithm for the temporal maintenance of a background model to enhance the rendering of occlusions and reduce temporal artefacts -LRB- flicker -RRB- ; and a cost aggregation algorithm that acts directly on our three-dimensional matching cost space .", "output": {"relations": {"used for": [{"head": {"text": "algorithm", "start": 42, "end": 51}, "tail": {"text": "temporal maintenance of a background model", "start": 60, "end": 102}}, {"head": {"text": "algorithm", "start": 42, "end": 51}, "tail": {"text": "rendering of occlusions", "start": 118, "end": 141}}, {"head": {"text": "algorithm", "start": 42, "end": 51}, "tail": {"text": "temporal artefacts -LRB- flicker -RRB-", "start": 153, "end": 191}}, {"head": {"text": "cost aggregation algorithm", "start": 200, "end": 226}, "tail": {"text": "three-dimensional matching cost space", "start": 253, "end": 290}}], "conjunction": [{"head": {"text": "cost aggregation algorithm", "start": 200, "end": 226}, "tail": {"text": "algorithm", "start": 42, "end": 51}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Examples are given that demonstrate the robustness of the new algorithm to spatial and temporal artefacts for long stereo video streams .", "output": {"relations": {"evaluate for": [{"head": {"text": "robustness", "start": 40, "end": 50}, "tail": {"text": "algorithm", "start": 62, "end": 71}}], "used for": [{"head": {"text": "algorithm", "start": 62, "end": 71}, "tail": {"text": "spatial and temporal artefacts", "start": 75, "end": 105}}, {"head": {"text": "spatial and temporal artefacts", "start": 75, "end": 105}, "tail": {"text": "long stereo video streams", "start": 110, "end": 135}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We further demonstrate synthesis from a freely translating virtual camera .", "output": {"relations": {"used for": [{"head": {"text": "translating virtual camera", "start": 47, "end": 73}, "tail": {"text": "synthesis", "start": 23, "end": 32}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "To a large extent , these statistics reflect semantic constraints and thus are used to disambiguate anaphora references and syntactic ambiguities .", "output": {"relations": {"used for": [{"head": {"text": "semantic constraints", "start": 45, "end": 65}, "tail": {"text": "anaphora references", "start": 100, "end": 119}}, {"head": {"text": "semantic constraints", "start": 45, "end": 65}, "tail": {"text": "syntactic ambiguities", "start": 124, "end": 145}}], "conjunction": [{"head": {"text": "anaphora references", "start": 100, "end": 119}, "tail": {"text": "syntactic ambiguities", "start": 124, "end": 145}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The results of the experiment show that in most of the cases the cooccurrence statistics indeed reflect the semantic constraints and thus provide a basis for a useful disambiguation tool .", "output": {"relations": {"used for": [{"head": {"text": "cooccurrence statistics", "start": 65, "end": 88}, "tail": {"text": "disambiguation tool", "start": 167, "end": 186}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We present a novel method for discovering parallel sentences in comparable , non-parallel corpora .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 19, "end": 25}, "tail": {"text": "discovering parallel sentences", "start": 30, "end": 60}}, {"head": {"text": "comparable , non-parallel corpora", "start": 64, "end": 97}, "tail": {"text": "discovering parallel sentences", "start": 30, "end": 60}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Using this approach , we extract parallel data from large Chinese , Arabic , and English non-parallel newspaper corpora .", "output": {"relations": {"used for": [{"head": {"text": "approach", "start": 11, "end": 19}, "tail": {"text": "parallel data", "start": 33, "end": 46}}], "part of": [{"head": {"text": "parallel data", "start": 33, "end": 46}, "tail": {"text": "Chinese , Arabic , and English non-parallel newspaper corpora", "start": 58, "end": 119}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We evaluate the quality of the extracted data by showing that it improves the performance of a state-of-the-art statistical machine translation system .", "output": {"relations": {"used for": [{"head": {"text": "it", "start": 20, "end": 22}, "tail": {"text": "statistical machine translation system", "start": 112, "end": 150}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We also show that a good-quality MT system can be built from scratch by starting with a very small parallel corpus -LRB- 100,000 words -RRB- and exploiting a large non-parallel corpus .", "output": {"relations": {"used for": [{"head": {"text": "parallel corpus", "start": 99, "end": 114}, "tail": {"text": "MT system", "start": 33, "end": 42}}, {"head": {"text": "non-parallel corpus", "start": 164, "end": 183}, "tail": {"text": "MT system", "start": 33, "end": 42}}], "conjunction": [{"head": {"text": "parallel corpus", "start": 99, "end": 114}, "tail": {"text": "non-parallel corpus", "start": 164, "end": 183}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Thus , our method can be applied with great benefit to language pairs for which only scarce resources are available .", "output": {"relations": {"used for": [{"head": {"text": "scarce resources", "start": 85, "end": 101}, "tail": {"text": "method", "start": 11, "end": 17}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , we describe a search procedure for statistical machine translation -LRB- MT -RRB- based on dynamic programming -LRB- DP -RRB- .", "output": {"relations": {"used for": [{"head": {"text": "search procedure", "start": 30, "end": 46}, "tail": {"text": "statistical machine translation -LRB- MT -RRB-", "start": 51, "end": 97}}, {"head": {"text": "dynamic programming -LRB- DP -RRB-", "start": 107, "end": 141}, "tail": {"text": "statistical machine translation -LRB- MT -RRB-", "start": 51, "end": 97}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Starting from a DP-based solution to the traveling salesman problem , we present a novel technique to restrict the possible word reordering between source and target language in order to achieve an efficient search algorithm .", "output": {"relations": {"used for": [{"head": {"text": "technique", "start": 89, "end": 98}, "tail": {"text": "search algorithm", "start": 208, "end": 224}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The experimental tests are carried out on the Verbmobil task -LRB- German-English , 8000-word vocabulary -RRB- , which is a limited-domain spoken-language task .", "output": {"relations": {"hyponym of": [{"head": {"text": "Verbmobil task", "start": 46, "end": 60}, "tail": {"text": "limited-domain spoken-language task", "start": 124, "end": 159}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A purely functional implementation of LR-parsers is given , together with a simple correctness proof .", "output": {"relations": {"conjunction": [{"head": {"text": "correctness proof", "start": 83, "end": 100}, "tail": {"text": "LR-parsers", "start": 38, "end": 48}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "It is presented as a generalization of the recursive descent parser .", "output": {"relations": {"used for": [{"head": {"text": "recursive descent parser", "start": 43, "end": 67}, "tail": {"text": "It", "start": 0, "end": 2}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "For non-LR grammars the time-complexity of our parser is cubic if the functions that constitute the parser are implemented as memo-functions , i.e. functions that memorize the results of previous invocations .", "output": {"relations": {"evaluate for": [{"head": {"text": "time-complexity", "start": 24, "end": 39}, "tail": {"text": "parser", "start": 47, "end": 53}}], "used for": [{"head": {"text": "parser", "start": 47, "end": 53}, "tail": {"text": "non-LR grammars", "start": 4, "end": 19}}, {"head": {"text": "memo-functions", "start": 126, "end": 140}, "tail": {"text": "parser", "start": 47, "end": 53}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Memo-functions also facilitate a simple way to construct a very compact representation of the parse forest .", "output": {"relations": {"used for": [{"head": {"text": "Memo-functions", "start": 0, "end": 14}, "tail": {"text": "parse forest", "start": 94, "end": 106}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "For LR -LRB- 0 -RRB- grammars , our algorithm is closely related to the recursive ascent parsers recently discovered by Kruse-man Aretz -LSB- 1 -RSB- and Roberts -LSB- 2 -RSB- .", "output": {"relations": {"used for": [{"head": {"text": "algorithm", "start": 36, "end": 45}, "tail": {"text": "LR -LRB- 0 -RRB- grammars", "start": 4, "end": 29}}], "conjunction": [{"head": {"text": "algorithm", "start": 36, "end": 45}, "tail": {"text": "recursive ascent parsers", "start": 72, "end": 96}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Extended CF grammars -LRB- grammars with regular expressions at the right hand side -RRB- can be parsed with a simple modification of the LR-parser for normal CF grammars .", "output": {"relations": {"feature of": [{"head": {"text": "regular expressions", "start": 41, "end": 60}, "tail": {"text": "grammars", "start": 12, "end": 20}}], "used for": [{"head": {"text": "LR-parser", "start": 138, "end": 147}, "tail": {"text": "Extended CF grammars", "start": 0, "end": 20}}, {"head": {"text": "LR-parser", "start": 138, "end": 147}, "tail": {"text": "CF grammars", "start": 9, "end": 20}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this theory , discourse structure is composed of three separate but interrelated components : the structure of the sequence of utterances -LRB- called the linguistic structure -RRB- , a structure of purposes -LRB- called the intentional structure -RRB- , and the state of focus of attention -LRB- called the attentional state -RRB- .", "output": {"relations": {"part of": [{"head": {"text": "components", "start": 84, "end": 94}, "tail": {"text": "discourse structure", "start": 17, "end": 36}}, {"head": {"text": "linguistic structure", "start": 158, "end": 178}, "tail": {"text": "components", "start": 84, "end": 94}}, {"head": {"text": "intentional structure", "start": 228, "end": 249}, "tail": {"text": "components", "start": 84, "end": 94}}, {"head": {"text": "attentional state", "start": 311, "end": 328}, "tail": {"text": "components", "start": 84, "end": 94}}], "conjunction": [{"head": {"text": "linguistic structure", "start": 158, "end": 178}, "tail": {"text": "intentional structure", "start": 228, "end": 249}}, {"head": {"text": "intentional structure", "start": 228, "end": 249}, "tail": {"text": "attentional state", "start": 311, "end": 328}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The intentional structure captures the discourse-relevant purposes , expressed in each of the linguistic segments as well as relationships among them .", "output": {"relations": {"used for": [{"head": {"text": "intentional structure", "start": 4, "end": 25}, "tail": {"text": "discourse-relevant purposes", "start": 39, "end": 66}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The distinction among these components is essential to provide an adequate explanation of such discourse phenomena as cue phrases , referring expressions , and interruptions .", "output": {"relations": {"hyponym of": [{"head": {"text": "cue phrases", "start": 118, "end": 129}, "tail": {"text": "discourse phenomena", "start": 95, "end": 114}}, {"head": {"text": "referring expressions", "start": 132, "end": 153}, "tail": {"text": "discourse phenomena", "start": 95, "end": 114}}, {"head": {"text": "interruptions", "start": 160, "end": 173}, "tail": {"text": "discourse phenomena", "start": 95, "end": 114}}], "conjunction": [{"head": {"text": "cue phrases", "start": 118, "end": 129}, "tail": {"text": "referring expressions", "start": 132, "end": 153}}, {"head": {"text": "referring expressions", "start": 132, "end": 153}, "tail": {"text": "interruptions", "start": 160, "end": 173}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We examine the relationship between the two grammatical formalisms : Tree Adjoining Grammars and Head Grammars .", "output": {"relations": {"hyponym of": [{"head": {"text": "Tree Adjoining Grammars", "start": 69, "end": 92}, "tail": {"text": "grammatical formalisms", "start": 44, "end": 66}}, {"head": {"text": "Head Grammars", "start": 97, "end": 110}, "tail": {"text": "grammatical formalisms", "start": 44, "end": 66}}], "compare": [{"head": {"text": "Tree Adjoining Grammars", "start": 69, "end": 92}, "tail": {"text": "Head Grammars", "start": 97, "end": 110}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We then turn to a discussion comparing the linguistic expressiveness of the two formalisms .", "output": {"relations": {"feature of": [{"head": {"text": "linguistic expressiveness", "start": 43, "end": 68}, "tail": {"text": "formalisms", "start": 80, "end": 90}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We provide a unified account of sentence-level and text-level anaphora within the framework of a dependency-based grammar model .", "output": {"relations": {"used for": [{"head": {"text": "dependency-based grammar model", "start": 97, "end": 127}, "tail": {"text": "sentence-level and text-level anaphora", "start": 32, "end": 70}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Criteria for anaphora resolution within sentence boundaries rephrase major concepts from GB 's binding theory , while those for text-level anaphora incorporate an adapted version of a Grosz-Sidner-style focus model .", "output": {"relations": {"used for": [{"head": {"text": "Criteria", "start": 0, "end": 8}, "tail": {"text": "anaphora resolution within sentence boundaries", "start": 13, "end": 59}}, {"head": {"text": "GB 's binding theory", "start": 89, "end": 109}, "tail": {"text": "Criteria", "start": 0, "end": 8}}, {"head": {"text": "those", "start": 118, "end": 123}, "tail": {"text": "text-level anaphora", "start": 128, "end": 147}}], "part of": [{"head": {"text": "Grosz-Sidner-style focus model", "start": 184, "end": 214}, "tail": {"text": "those", "start": 118, "end": 123}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Coedition of a natural language text and its representation in some interlingual form seems the best and simplest way to share text revision across languages .", "output": {"relations": {"used for": [{"head": {"text": "Coedition", "start": 0, "end": 9}, "tail": {"text": "text revision", "start": 127, "end": 140}}, {"head": {"text": "natural language text", "start": 15, "end": 36}, "tail": {"text": "Coedition", "start": 0, "end": 9}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The modified graph is then sent to the UNL-L0 deconverter and the result shown .", "output": {"relations": {"used for": [{"head": {"text": "graph", "start": 13, "end": 18}, "tail": {"text": "UNL-L0 deconverter", "start": 39, "end": 57}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "On the internal side , liaisons are established between elements of the text and the graph by using broadly available resources such as a LO-English or better a L0-UNL dictionary , a morphosyntactic parser of L0 , and a canonical graph2tree transformation .", "output": {"relations": {"used for": [{"head": {"text": "resources", "start": 118, "end": 127}, "tail": {"text": "liaisons", "start": 23, "end": 31}}], "hyponym of": [{"head": {"text": "LO-English or better a L0-UNL dictionary", "start": 138, "end": 178}, "tail": {"text": "resources", "start": 118, "end": 127}}, {"head": {"text": "morphosyntactic parser of L0", "start": 183, "end": 211}, "tail": {"text": "resources", "start": 118, "end": 127}}, {"head": {"text": "canonical graph2tree transformation", "start": 220, "end": 255}, "tail": {"text": "resources", "start": 118, "end": 127}}], "conjunction": [{"head": {"text": "LO-English or better a L0-UNL dictionary", "start": 138, "end": 178}, "tail": {"text": "morphosyntactic parser of L0", "start": 183, "end": 211}}, {"head": {"text": "morphosyntactic parser of L0", "start": 183, "end": 211}, "tail": {"text": "canonical graph2tree transformation", "start": 220, "end": 255}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Establishing a `` best '' correspondence between the '' UNL-tree + L0 '' and the '' MS-L0 structure '' , a lattice , may be done using the dictionary and trying to align the tree and the selected trajectory with as few crossing liaisons as possible .", "output": {"relations": {"conjunction": [{"head": {"text": "UNL-tree + L0", "start": 56, "end": 69}, "tail": {"text": "MS-L0 structure", "start": 84, "end": 99}}], "used for": [{"head": {"text": "dictionary", "start": 139, "end": 149}, "tail": {"text": "lattice", "start": 107, "end": 114}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A central goal of this research is to merge approaches from pivot MT , interactive MT , and multilingual text authoring .", "output": {"relations": {"conjunction": [{"head": {"text": "pivot MT", "start": 60, "end": 68}, "tail": {"text": "interactive MT", "start": 71, "end": 85}}, {"head": {"text": "interactive MT", "start": 71, "end": 85}, "tail": {"text": "multilingual text authoring", "start": 92, "end": 119}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We report experiments conducted on a multilingual corpus to estimate the number of analogies among the sentences that it contains .", "output": {"relations": {"evaluate for": [{"head": {"text": "multilingual corpus", "start": 37, "end": 56}, "tail": {"text": "analogies", "start": 83, "end": 92}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our goal is to learn a Mahalanobis distance by minimizing a loss defined on the weighted sum of the precision at different ranks .", "output": {"relations": {"used for": [{"head": {"text": "loss", "start": 60, "end": 64}, "tail": {"text": "Mahalanobis distance", "start": 23, "end": 43}}], "feature of": [{"head": {"text": "weighted sum", "start": 80, "end": 92}, "tail": {"text": "precision", "start": 100, "end": 109}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our core motivation is that minimizing a weighted rank loss is a natural criterion for many problems in computer vision such as person re-identification .", "output": {"relations": {"used for": [{"head": {"text": "weighted rank loss", "start": 41, "end": 59}, "tail": {"text": "computer vision", "start": 104, "end": 119}}, {"head": {"text": "weighted rank loss", "start": 41, "end": 59}, "tail": {"text": "person re-identification", "start": 128, "end": 152}}], "hyponym of": [{"head": {"text": "person re-identification", "start": 128, "end": 152}, "tail": {"text": "computer vision", "start": 104, "end": 119}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We propose a novel metric learning formulation called Weighted Approximate Rank Component Analysis -LRB- WARCA -RRB- .", "output": {"relations": {"hyponym of": [{"head": {"text": "Weighted Approximate Rank Component Analysis -LRB- WARCA -RRB-", "start": 54, "end": 116}, "tail": {"text": "metric learning formulation", "start": 19, "end": 46}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We then derive a scalable stochastic gradient descent algorithm for the resulting learning problem .", "output": {"relations": {"used for": [{"head": {"text": "stochastic gradient descent algorithm", "start": 26, "end": 63}, "tail": {"text": "learning problem", "start": 82, "end": 98}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We also derive an efficient non-linear extension of WARCA by using the kernel trick .", "output": {"relations": {"used for": [{"head": {"text": "kernel trick", "start": 71, "end": 83}, "tail": {"text": "non-linear extension of WARCA", "start": 28, "end": 57}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Kernel space embedding decouples the training and prediction costs from the data dimension and enables us to plug inarbitrary distance measures which are more natural for the features .", "output": {"relations": {"used for": [{"head": {"text": "Kernel space embedding", "start": 0, "end": 22}, "tail": {"text": "inarbitrary distance measures", "start": 114, "end": 143}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We also address a more general problem of matrix rank degeneration & non-isolated minima in the low-rank matrix optimization by using new type of regularizer which approximately enforces the or-thonormality of the learned matrix very efficiently .", "output": {"relations": {"conjunction": [{"head": {"text": "matrix rank degeneration", "start": 42, "end": 66}, "tail": {"text": "non-isolated minima", "start": 69, "end": 88}}], "feature of": [{"head": {"text": "matrix rank degeneration", "start": 42, "end": 66}, "tail": {"text": "low-rank matrix optimization", "start": 96, "end": 124}}, {"head": {"text": "non-isolated minima", "start": 69, "end": 88}, "tail": {"text": "low-rank matrix optimization", "start": 96, "end": 124}}, {"head": {"text": "or-thonormality", "start": 191, "end": 206}, "tail": {"text": "learned matrix", "start": 214, "end": 228}}], "used for": [{"head": {"text": "regularizer", "start": 146, "end": 157}, "tail": {"text": "low-rank matrix optimization", "start": 96, "end": 124}}, {"head": {"text": "regularizer", "start": 146, "end": 157}, "tail": {"text": "or-thonormality", "start": 191, "end": 206}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We validate this new method on nine standard person re-identification datasets including two large scale Market-1501 and CUHK03 datasets and show that we improve upon the current state-of-the-art methods on all of them .", "output": {"relations": {"evaluate for": [{"head": {"text": "person re-identification datasets", "start": 45, "end": 78}, "tail": {"text": "method", "start": 21, "end": 27}}], "hyponym of": [{"head": {"text": "scale Market-1501", "start": 99, "end": 116}, "tail": {"text": "person re-identification datasets", "start": 45, "end": 78}}, {"head": {"text": "CUHK03 datasets", "start": 121, "end": 136}, "tail": {"text": "person re-identification datasets", "start": 45, "end": 78}}], "conjunction": [{"head": {"text": "CUHK03 datasets", "start": 121, "end": 136}, "tail": {"text": "scale Market-1501", "start": 99, "end": 116}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , we discuss language model adaptation methods given a word list and a raw corpus .", "output": {"relations": {"used for": [{"head": {"text": "word list", "start": 69, "end": 78}, "tail": {"text": "language model adaptation methods", "start": 27, "end": 60}}, {"head": {"text": "raw corpus", "start": 85, "end": 95}, "tail": {"text": "language model adaptation methods", "start": 27, "end": 60}}], "conjunction": [{"head": {"text": "word list", "start": 69, "end": 78}, "tail": {"text": "raw corpus", "start": 85, "end": 95}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this situation , the general method is to segment the raw corpus automatically using a word list , correct the output sentences by hand , and build a model from the segmented corpus .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 32, "end": 38}, "tail": {"text": "raw corpus", "start": 57, "end": 67}}, {"head": {"text": "word list", "start": 90, "end": 99}, "tail": {"text": "method", "start": 32, "end": 38}}, {"head": {"text": "segmented corpus", "start": 168, "end": 184}, "tail": {"text": "model", "start": 153, "end": 158}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In the experiments , we used a variety of methods for preparing a segmented corpus and compared the language models by their speech recognition accuracies .", "output": {"relations": {"used for": [{"head": {"text": "methods", "start": 42, "end": 49}, "tail": {"text": "preparing a segmented corpus", "start": 54, "end": 82}}], "evaluate for": [{"head": {"text": "speech recognition accuracies", "start": 125, "end": 154}, "tail": {"text": "language models", "start": 100, "end": 115}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Many practical modeling problems involve discrete data that are best represented as draws from multinomial or categorical distributions .", "output": {"relations": {"used for": [{"head": {"text": "discrete data", "start": 41, "end": 54}, "tail": {"text": "modeling problems", "start": 15, "end": 32}}, {"head": {"text": "multinomial or categorical distributions", "start": 95, "end": 135}, "tail": {"text": "modeling problems", "start": 15, "end": 32}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "For example , nucleotides in a DNA sequence , children 's names in a given state and year , and text documents are all commonly modeled with multinomial distributions .", "output": {"relations": {"used for": [{"head": {"text": "multinomial distributions", "start": 141, "end": 166}, "tail": {"text": "nucleotides in a DNA sequence", "start": 14, "end": 43}}, {"head": {"text": "multinomial distributions", "start": 141, "end": 166}, "tail": {"text": "text documents", "start": 96, "end": 110}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Here , we leverage a logistic stick-breaking representation and recent innovations in Pólya-gamma augmentation to reformu-late the multinomial distribution in terms of latent variables with jointly Gaussian likelihoods , enabling us to take advantage of a host of Bayesian inference techniques for Gaussian models with minimal overhead .", "output": {"relations": {"used for": [{"head": {"text": "logistic stick-breaking representation", "start": 21, "end": 59}, "tail": {"text": "multinomial distribution", "start": 131, "end": 155}}, {"head": {"text": "Pólya-gamma augmentation", "start": 86, "end": 110}, "tail": {"text": "multinomial distribution", "start": 131, "end": 155}}, {"head": {"text": "Bayesian inference techniques", "start": 264, "end": 293}, "tail": {"text": "Gaussian models", "start": 298, "end": 313}}], "part of": [{"head": {"text": "latent variables", "start": 168, "end": 184}, "tail": {"text": "multinomial distribution", "start": 131, "end": 155}}], "feature of": [{"head": {"text": "jointly Gaussian likelihoods", "start": 190, "end": 218}, "tail": {"text": "latent variables", "start": 168, "end": 184}}, {"head": {"text": "minimal overhead", "start": 319, "end": 335}, "tail": {"text": "Gaussian models", "start": 298, "end": 313}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "MINPRAN , a new robust operator , nds good ts in data sets where more than 50 % of the points are outliers .", "output": {"relations": {"hyponym of": [{"head": {"text": "MINPRAN", "start": 0, "end": 7}, "tail": {"text": "robust operator", "start": 16, "end": 31}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Unlike other techniques that handle large outlier percentages , MINPRAN does not rely on a known error bound for the good data .", "output": {"relations": {"used for": [{"head": {"text": "techniques", "start": 13, "end": 23}, "tail": {"text": "large outlier percentages", "start": 36, "end": 61}}], "compare": [{"head": {"text": "techniques", "start": 13, "end": 23}, "tail": {"text": "MINPRAN", "start": 64, "end": 71}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Based on this , MINPRAN uses random sampling to search for the t and the number of inliers to the t that are least likely to have occurred randomly .", "output": {"relations": {"used for": [{"head": {"text": "random sampling", "start": 29, "end": 44}, "tail": {"text": "MINPRAN", "start": 16, "end": 23}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "MINPRAN 's properties are connrmed experimentally on synthetic data and compare favorably to least median of squares .", "output": {"relations": {"evaluate for": [{"head": {"text": "synthetic data", "start": 53, "end": 67}, "tail": {"text": "MINPRAN", "start": 0, "end": 7}}], "compare": [{"head": {"text": "least median of squares", "start": 93, "end": 116}, "tail": {"text": "MINPRAN", "start": 0, "end": 7}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Related work applies MINPRAN to complex range and intensity data 23 -RSB- .", "output": {"relations": {"used for": [{"head": {"text": "MINPRAN", "start": 21, "end": 28}, "tail": {"text": "complex range", "start": 32, "end": 45}}, {"head": {"text": "MINPRAN", "start": 21, "end": 28}, "tail": {"text": "intensity data", "start": 50, "end": 64}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Metagrammatical formalisms that combine context-free phrase structure rules and metarules -LRB- MPS grammars -RRB- allow concise statement of generalizations about the syntax of natural languages .", "output": {"relations": {"part of": [{"head": {"text": "context-free phrase structure rules", "start": 40, "end": 75}, "tail": {"text": "Metagrammatical formalisms", "start": 0, "end": 26}}, {"head": {"text": "metarules -LRB- MPS grammars -RRB-", "start": 80, "end": 114}, "tail": {"text": "Metagrammatical formalisms", "start": 0, "end": 26}}], "conjunction": [{"head": {"text": "context-free phrase structure rules", "start": 40, "end": 75}, "tail": {"text": "metarules -LRB- MPS grammars -RRB-", "start": 80, "end": 114}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We evaluate several proposals for constraining them , basing our assessment on computational tractability and explanatory adequacy .", "output": {"relations": {"evaluate for": [{"head": {"text": "computational tractability and explanatory adequacy", "start": 79, "end": 130}, "tail": {"text": "them", "start": 47, "end": 51}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The unique properties of tree-adjoining grammars -LRB- TAG -RRB- present a challenge for the application of TAGs beyond the limited confines of syntax , for instance , to the task of semantic interpretation or automatic translation of natural language .", "output": {"relations": {"used for": [{"head": {"text": "TAGs", "start": 108, "end": 112}, "tail": {"text": "semantic interpretation", "start": 183, "end": 206}}, {"head": {"text": "TAGs", "start": 108, "end": 112}, "tail": {"text": "automatic translation of natural language", "start": 210, "end": 251}}], "conjunction": [{"head": {"text": "semantic interpretation", "start": 183, "end": 206}, "tail": {"text": "automatic translation of natural language", "start": 210, "end": 251}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The formalism 's intended usage is to relate expressions of natural languages to their associated semantics represented in a logical form language , or to their translates in another natural language ; in summary , we intend it to allow TAGs to be used beyond their role in syntax proper .", "output": {"relations": {"used for": [{"head": {"text": "logical form language", "start": 125, "end": 146}, "tail": {"text": "semantics", "start": 98, "end": 107}}, {"head": {"text": "TAGs", "start": 237, "end": 241}, "tail": {"text": "syntax proper", "start": 274, "end": 287}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A model-based approach to on-line cursive handwriting analysis and recognition is presented and evaluated .", "output": {"relations": {"used for": [{"head": {"text": "model-based approach", "start": 2, "end": 22}, "tail": {"text": "on-line cursive handwriting analysis and recognition", "start": 26, "end": 78}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this model , on-line handwriting is considered as a modulation of a simple cycloidal pen motion , described by two coupled oscillations with a constant linear drift along the line of the writing .", "output": {"relations": {"used for": [{"head": {"text": "model", "start": 8, "end": 13}, "tail": {"text": "on-line handwriting", "start": 16, "end": 35}}], "part of": [{"head": {"text": "on-line handwriting", "start": 16, "end": 35}, "tail": {"text": "cycloidal pen motion", "start": 78, "end": 98}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A general procedure for the estimation and quantization of these cycloidal motion parameters for arbitrary handwriting is presented .", "output": {"relations": {"used for": [{"head": {"text": "cycloidal motion parameters", "start": 65, "end": 92}, "tail": {"text": "arbitrary handwriting", "start": 97, "end": 118}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The result is a discrete motor control representation of the continuous pen motion , via the quantized levels of the model parameters .", "output": {"relations": {"used for": [{"head": {"text": "discrete motor control representation", "start": 16, "end": 53}, "tail": {"text": "continuous pen motion", "start": 61, "end": 82}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This motor control representation enables successful word spotting and matching of cursive scripts .", "output": {"relations": {"used for": [{"head": {"text": "motor control representation", "start": 5, "end": 33}, "tail": {"text": "word spotting", "start": 53, "end": 66}}, {"head": {"text": "motor control representation", "start": 5, "end": 33}, "tail": {"text": "matching of cursive scripts", "start": 71, "end": 98}}], "conjunction": [{"head": {"text": "word spotting", "start": 53, "end": 66}, "tail": {"text": "matching of cursive scripts", "start": 71, "end": 98}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our experiments clearly indicate the potential of this dynamic representation for complete cursive handwriting recognition .", "output": {"relations": {"used for": [{"head": {"text": "dynamic representation", "start": 55, "end": 77}, "tail": {"text": "cursive handwriting recognition", "start": 91, "end": 122}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In the Object Recognition task , there exists a di-chotomy between the categorization of objects and estimating object pose , where the former necessitates a view-invariant representation , while the latter requires a representation capable of capturing pose information over different categories of objects .", "output": {"relations": {"part of": [{"head": {"text": "categorization of objects", "start": 71, "end": 96}, "tail": {"text": "Object Recognition task", "start": 7, "end": 30}}, {"head": {"text": "estimating object pose", "start": 101, "end": 123}, "tail": {"text": "Object Recognition task", "start": 7, "end": 30}}], "conjunction": [{"head": {"text": "categorization of objects", "start": 71, "end": 96}, "tail": {"text": "estimating object pose", "start": 101, "end": 123}}], "used for": [{"head": {"text": "view-invariant representation", "start": 158, "end": 187}, "tail": {"text": "former", "start": 136, "end": 142}}, {"head": {"text": "representation", "start": 173, "end": 187}, "tail": {"text": "latter", "start": 200, "end": 206}}, {"head": {"text": "representation", "start": 173, "end": 187}, "tail": {"text": "pose information", "start": 254, "end": 270}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "With the rise of deep archi-tectures , the prime focus has been on object category recognition .", "output": {"relations": {"used for": [{"head": {"text": "deep archi-tectures", "start": 17, "end": 36}, "tail": {"text": "object category recognition", "start": 67, "end": 94}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In contrast , object pose estimation using these approaches has received relatively less attention .", "output": {"relations": {"used for": [{"head": {"text": "approaches", "start": 49, "end": 59}, "tail": {"text": "object pose estimation", "start": 14, "end": 36}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this work , we study how Convolutional Neural Networks -LRB- CNN -RRB- architectures can be adapted to the task of simultaneous object recognition and pose estimation .", "output": {"relations": {"used for": [{"head": {"text": "Convolutional Neural Networks -LRB- CNN -RRB- architectures", "start": 28, "end": 87}, "tail": {"text": "object recognition", "start": 131, "end": 149}}, {"head": {"text": "Convolutional Neural Networks -LRB- CNN -RRB- architectures", "start": 28, "end": 87}, "tail": {"text": "pose estimation", "start": 154, "end": 169}}], "conjunction": [{"head": {"text": "object recognition", "start": 131, "end": 149}, "tail": {"text": "pose estimation", "start": 154, "end": 169}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We investigate and analyze the layers of various CNN models and extensively compare between them with the goal of discovering how the layers of distributed representations within CNNs represent object pose information and how this contradicts with object category representations .", "output": {"relations": {"part of": [{"head": {"text": "layers", "start": 31, "end": 37}, "tail": {"text": "CNN models", "start": 49, "end": 59}}, {"head": {"text": "layers of distributed representations", "start": 134, "end": 171}, "tail": {"text": "CNNs", "start": 179, "end": 183}}], "used for": [{"head": {"text": "layers of distributed representations", "start": 134, "end": 171}, "tail": {"text": "object pose information", "start": 194, "end": 217}}], "compare": [{"head": {"text": "this", "start": 226, "end": 230}, "tail": {"text": "object category representations", "start": 248, "end": 279}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "It is particularly valuable to empirical MT research .", "output": {"relations": {"used for": [{"head": {"text": "It", "start": 0, "end": 2}, "tail": {"text": "empirical MT research", "start": 31, "end": 52}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , we explore geometric structures of 3D lines in ray space for improving light field triangulation and stereo matching .", "output": {"relations": {"used for": [{"head": {"text": "geometric structures of 3D lines", "start": 27, "end": 59}, "tail": {"text": "light field triangulation", "start": 87, "end": 112}}, {"head": {"text": "geometric structures of 3D lines", "start": 27, "end": 59}, "tail": {"text": "stereo matching", "start": 117, "end": 132}}], "feature of": [{"head": {"text": "ray space", "start": 63, "end": 72}, "tail": {"text": "geometric structures of 3D lines", "start": 27, "end": 59}}], "conjunction": [{"head": {"text": "light field triangulation", "start": 87, "end": 112}, "tail": {"text": "stereo matching", "start": 117, "end": 132}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Such a triangulation provides a piecewise-linear interpolant useful for light field super-resolution .", "output": {"relations": {"used for": [{"head": {"text": "triangulation", "start": 7, "end": 20}, "tail": {"text": "piecewise-linear interpolant", "start": 32, "end": 60}}, {"head": {"text": "piecewise-linear interpolant", "start": 32, "end": 60}, "tail": {"text": "light field super-resolution", "start": 72, "end": 100}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Experiments on synthetic and real data show that both our triangulation and LAGC algorithms outperform state-of-the-art solutions in accuracy and visual quality .", "output": {"relations": {"evaluate for": [{"head": {"text": "synthetic and real data", "start": 15, "end": 38}, "tail": {"text": "triangulation and LAGC algorithms", "start": 58, "end": 91}}, {"head": {"text": "synthetic and real data", "start": 15, "end": 38}, "tail": {"text": "state-of-the-art solutions", "start": 103, "end": 129}}, {"head": {"text": "accuracy", "start": 133, "end": 141}, "tail": {"text": "triangulation and LAGC algorithms", "start": 58, "end": 91}}, {"head": {"text": "accuracy", "start": 133, "end": 141}, "tail": {"text": "state-of-the-art solutions", "start": 103, "end": 129}}, {"head": {"text": "visual quality", "start": 146, "end": 160}, "tail": {"text": "triangulation and LAGC algorithms", "start": 58, "end": 91}}, {"head": {"text": "visual quality", "start": 146, "end": 160}, "tail": {"text": "state-of-the-art solutions", "start": 103, "end": 129}}], "compare": [{"head": {"text": "triangulation and LAGC algorithms", "start": 58, "end": 91}, "tail": {"text": "state-of-the-art solutions", "start": 103, "end": 129}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper presents a phrase-based statistical machine translation method , based on non-contiguous phrases , i.e. phrases with gaps .", "output": {"relations": {"used for": [{"head": {"text": "non-contiguous phrases", "start": 85, "end": 107}, "tail": {"text": "phrase-based statistical machine translation method", "start": 22, "end": 73}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A method for producing such phrases from a word-aligned corpora is proposed .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 2, "end": 8}, "tail": {"text": "phrases", "start": 28, "end": 35}}], "evaluate for": [{"head": {"text": "word-aligned corpora", "start": 43, "end": 63}, "tail": {"text": "method", "start": 2, "end": 8}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A statistical translation model is also presented that deals such phrases , as well as a training method based on the maximization of translation accuracy , as measured with the NIST evaluation metric .", "output": {"relations": {"used for": [{"head": {"text": "statistical translation model", "start": 2, "end": 31}, "tail": {"text": "phrases", "start": 66, "end": 73}}, {"head": {"text": "maximization of translation accuracy", "start": 118, "end": 154}, "tail": {"text": "training method", "start": 89, "end": 104}}], "evaluate for": [{"head": {"text": "NIST evaluation metric", "start": 178, "end": 200}, "tail": {"text": "statistical translation model", "start": 2, "end": 31}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Translations are produced by means of a beam-search decoder .", "output": {"relations": {"used for": [{"head": {"text": "beam-search decoder", "start": 40, "end": 59}, "tail": {"text": "Translations", "start": 0, "end": 12}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "GLOSSER is designed to support reading and learning to read in a foreign language .", "output": {"relations": {"used for": [{"head": {"text": "GLOSSER", "start": 0, "end": 7}, "tail": {"text": "reading and learning", "start": 31, "end": 51}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "There are four language pairs currently supported by GLOSSER : English-Bulgarian , English-Estonian , English-Hungarian and French-Dutch .", "output": {"relations": {"used for": [{"head": {"text": "language pairs", "start": 15, "end": 29}, "tail": {"text": "GLOSSER", "start": 53, "end": 60}}], "hyponym of": [{"head": {"text": "English-Bulgarian", "start": 63, "end": 80}, "tail": {"text": "language pairs", "start": 15, "end": 29}}, {"head": {"text": "English-Estonian", "start": 83, "end": 99}, "tail": {"text": "language pairs", "start": 15, "end": 29}}, {"head": {"text": "English-Hungarian", "start": 102, "end": 119}, "tail": {"text": "language pairs", "start": 15, "end": 29}}, {"head": {"text": "French-Dutch", "start": 124, "end": 136}, "tail": {"text": "language pairs", "start": 15, "end": 29}}], "conjunction": [{"head": {"text": "English-Bulgarian", "start": 63, "end": 80}, "tail": {"text": "English-Estonian", "start": 83, "end": 99}}, {"head": {"text": "English-Estonian", "start": 83, "end": 99}, "tail": {"text": "English-Hungarian", "start": 102, "end": 119}}, {"head": {"text": "English-Hungarian", "start": 102, "end": 119}, "tail": {"text": "French-Dutch", "start": 124, "end": 136}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A demonstration -LRB- in UNIX -RRB- for Applied Natural Language Processing emphasizes components put to novel technical uses in intelligent computer-assisted morphological analysis -LRB- ICALL -RRB- , including disambiguated morphological analysis and lemmatized indexing for an aligned bilingual corpus of word examples .", "output": {"relations": {"used for": [{"head": {"text": "components", "start": 87, "end": 97}, "tail": {"text": "intelligent computer-assisted morphological analysis -LRB- ICALL -RRB-", "start": 129, "end": 199}}, {"head": {"text": "disambiguated morphological analysis", "start": 212, "end": 248}, "tail": {"text": "aligned bilingual corpus", "start": 280, "end": 304}}, {"head": {"text": "lemmatized indexing", "start": 253, "end": 272}, "tail": {"text": "aligned bilingual corpus", "start": 280, "end": 304}}], "hyponym of": [{"head": {"text": "disambiguated morphological analysis", "start": 212, "end": 248}, "tail": {"text": "components", "start": 87, "end": 97}}, {"head": {"text": "lemmatized indexing", "start": 253, "end": 272}, "tail": {"text": "components", "start": 87, "end": 97}}], "conjunction": [{"head": {"text": "disambiguated morphological analysis", "start": 212, "end": 248}, "tail": {"text": "lemmatized indexing", "start": 253, "end": 272}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We present a new part-of-speech tagger that demonstrates the following ideas : -LRB- i -RRB- explicit use of both preceding and following tag contexts via a dependency network representation , -LRB- ii -RRB- broad use of lexical features , including jointly conditioning on multiple consecutive words , -LRB- iii -RRB- effective use of priors in conditional loglinear models , and -LRB- iv -RRB- fine-grained modeling of unknown word features .", "output": {"relations": {"used for": [{"head": {"text": "tag contexts", "start": 138, "end": 150}, "tail": {"text": "part-of-speech tagger", "start": 17, "end": 38}}, {"head": {"text": "dependency network representation", "start": 157, "end": 190}, "tail": {"text": "tag contexts", "start": 138, "end": 150}}, {"head": {"text": "lexical features", "start": 221, "end": 237}, "tail": {"text": "part-of-speech tagger", "start": 17, "end": 38}}, {"head": {"text": "priors in conditional loglinear models", "start": 336, "end": 374}, "tail": {"text": "part-of-speech tagger", "start": 17, "end": 38}}, {"head": {"text": "fine-grained modeling of unknown word features", "start": 396, "end": 442}, "tail": {"text": "part-of-speech tagger", "start": 17, "end": 38}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Using these ideas together , the resulting tagger gives a 97.24 % accuracy on the Penn Treebank WSJ , an error reduction of 4.4 % on the best previous single automatically learned tagging result .", "output": {"relations": {"evaluate for": [{"head": {"text": "accuracy", "start": 66, "end": 74}, "tail": {"text": "tagger", "start": 43, "end": 49}}, {"head": {"text": "Penn Treebank WSJ", "start": 82, "end": 99}, "tail": {"text": "tagger", "start": 43, "end": 49}}, {"head": {"text": "error", "start": 105, "end": 110}, "tail": {"text": "tagger", "start": 43, "end": 49}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Owing to these variations , the pedestrian data is distributed as highly-curved manifolds in the feature space , despite the current convolutional neural networks -LRB- CNN -RRB- 's capability of feature extraction .", "output": {"relations": {"used for": [{"head": {"text": "highly-curved manifolds", "start": 66, "end": 89}, "tail": {"text": "pedestrian data", "start": 32, "end": 47}}, {"head": {"text": "convolutional neural networks -LRB- CNN -RRB-", "start": 133, "end": 178}, "tail": {"text": "feature extraction", "start": 196, "end": 214}}], "feature of": [{"head": {"text": "feature space", "start": 97, "end": 110}, "tail": {"text": "highly-curved manifolds", "start": 66, "end": 89}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In practice , the current deep embedding methods use the Euclidean distance for the training and test .", "output": {"relations": {"used for": [{"head": {"text": "Euclidean distance", "start": 57, "end": 75}, "tail": {"text": "deep embedding methods", "start": 26, "end": 48}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "On the other hand , the manifold learning methods suggest to use the Euclidean distance in the local range , combining with the graphical relationship between samples , for approximating the geodesic distance .", "output": {"relations": {"used for": [{"head": {"text": "Euclidean distance", "start": 69, "end": 87}, "tail": {"text": "manifold learning methods", "start": 24, "end": 49}}, {"head": {"text": "Euclidean distance", "start": 69, "end": 87}, "tail": {"text": "geodesic distance", "start": 191, "end": 208}}, {"head": {"text": "graphical relationship", "start": 128, "end": 150}, "tail": {"text": "geodesic distance", "start": 191, "end": 208}}], "conjunction": [{"head": {"text": "Euclidean distance", "start": 69, "end": 87}, "tail": {"text": "graphical relationship", "start": 128, "end": 150}}], "feature of": [{"head": {"text": "local range", "start": 95, "end": 106}, "tail": {"text": "Euclidean distance", "start": 69, "end": 87}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "From this point of view , selecting suitable positive -LRB- i.e. intra-class -RRB- training samples within a local range is critical for training the CNN embedding , especially when the data has large intra-class variations .", "output": {"relations": {"feature of": [{"head": {"text": "intra-class variations", "start": 201, "end": 223}, "tail": {"text": "data", "start": 186, "end": 190}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , we propose a novel moderate positive sample mining method to train robust CNN for person re-identification , dealing with the problem of large variation .", "output": {"relations": {"used for": [{"head": {"text": "moderate positive sample mining method", "start": 35, "end": 73}, "tail": {"text": "robust CNN", "start": 83, "end": 93}}, {"head": {"text": "robust CNN", "start": 83, "end": 93}, "tail": {"text": "person re-identification", "start": 98, "end": 122}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In addition , we improve the learning by a metric weight constraint , so that the learned metric has a better generalization ability .", "output": {"relations": {"used for": [{"head": {"text": "metric weight constraint", "start": 43, "end": 67}, "tail": {"text": "learning", "start": 29, "end": 37}}], "feature of": [{"head": {"text": "generalization ability", "start": 110, "end": 132}, "tail": {"text": "learned metric", "start": 82, "end": 96}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Experiments show that these two strategies are effective in learning robust deep metrics for person re-identification , and accordingly our deep model significantly outperforms the state-of-the-art methods on several benchmarks of person re-identification .", "output": {"relations": {"used for": [{"head": {"text": "robust deep metrics", "start": 69, "end": 88}, "tail": {"text": "person re-identification", "start": 93, "end": 117}}, {"head": {"text": "deep model", "start": 140, "end": 150}, "tail": {"text": "person re-identification", "start": 93, "end": 117}}, {"head": {"text": "state-of-the-art methods", "start": 181, "end": 205}, "tail": {"text": "person re-identification", "start": 93, "end": 117}}], "compare": [{"head": {"text": "deep model", "start": 140, "end": 150}, "tail": {"text": "state-of-the-art methods", "start": 181, "end": 205}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Therefore , the study presented in this paper may be useful in inspiring new designs of deep models for person re-identification .", "output": {"relations": {"used for": [{"head": {"text": "deep models", "start": 88, "end": 99}, "tail": {"text": "person re-identification", "start": 104, "end": 128}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Utterance Verification -LRB- UV -RRB- is a critical function of an Automatic Speech Recognition -LRB- ASR -RRB- System working on real applications where spontaneous speech , out-of-vocabulary -LRB- OOV -RRB- words and acoustic noises are present .", "output": {"relations": {"hyponym of": [{"head": {"text": "Utterance Verification -LRB- UV -RRB-", "start": 0, "end": 37}, "tail": {"text": "Automatic Speech Recognition -LRB- ASR -RRB- System", "start": 67, "end": 118}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper we present a new UV procedure with two major features : a -RRB- Confidence tests are applied to decoded string hypotheses obtained from using word and garbage models that represent OOV words and noises .", "output": {"relations": {"used for": [{"head": {"text": "Confidence tests", "start": 78, "end": 94}, "tail": {"text": "decoded string hypotheses", "start": 110, "end": 135}}], "conjunction": [{"head": {"text": "noises", "start": 209, "end": 215}, "tail": {"text": "OOV words", "start": 195, "end": 204}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Thus the ASR system is designed to deal with what we refer to as Word Spotting and Noise Spotting capabilities .", "output": {"relations": {"used for": [{"head": {"text": "ASR system", "start": 9, "end": 19}, "tail": {"text": "Word Spotting", "start": 65, "end": 78}}, {"head": {"text": "ASR system", "start": 9, "end": 19}, "tail": {"text": "Noise Spotting capabilities", "start": 83, "end": 110}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "b -RRB- The UV procedure is based on three different confidence tests , two based on acoustic measures and one founded on linguistic information , applied in a hierarchical structure .", "output": {"relations": {"used for": [{"head": {"text": "confidence tests", "start": 53, "end": 69}, "tail": {"text": "UV procedure", "start": 12, "end": 24}}, {"head": {"text": "confidence tests", "start": 53, "end": 69}, "tail": {"text": "hierarchical structure", "start": 160, "end": 182}}, {"head": {"text": "acoustic measures", "start": 85, "end": 102}, "tail": {"text": "two", "start": 72, "end": 75}}, {"head": {"text": "linguistic information", "start": 122, "end": 144}, "tail": {"text": "one", "start": 107, "end": 110}}], "hyponym of": [{"head": {"text": "two", "start": 72, "end": 75}, "tail": {"text": "confidence tests", "start": 53, "end": 69}}, {"head": {"text": "one", "start": 107, "end": 110}, "tail": {"text": "confidence tests", "start": 53, "end": 69}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Experimental results from a real telephone application on a natural number recognition task show an 50 % reduction in recognition errors with a moderate 12 % rejection rate of correct utterances and a low 1.5 % rate of false acceptance .", "output": {"relations": {"feature of": [{"head": {"text": "natural number recognition task", "start": 60, "end": 91}, "tail": {"text": "telephone application", "start": 33, "end": 54}}], "evaluate for": [{"head": {"text": "recognition errors", "start": 118, "end": 136}, "tail": {"text": "natural number recognition task", "start": 60, "end": 91}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "A critical step in encoding sound for neuronal processing occurs when the analog pressure wave is coded into discrete nerve-action potentials .", "output": {"relations": {"used for": [{"head": {"text": "encoding sound", "start": 19, "end": 33}, "tail": {"text": "neuronal processing", "start": 38, "end": 57}}, {"head": {"text": "discrete nerve-action potentials", "start": 109, "end": 141}, "tail": {"text": "analog pressure wave", "start": 74, "end": 94}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Recent pool models of the inner hair cell synapse do not reproduce the dead time period after an intense stimulus , so we used visual inspection and automatic speech recognition -LRB- ASR -RRB- to investigate an offset adaptation -LRB- OA -RRB- model proposed by Zhang et al. -LSB- 1 -RSB- .", "output": {"relations": {"used for": [{"head": {"text": "pool models", "start": 7, "end": 18}, "tail": {"text": "inner hair cell synapse", "start": 26, "end": 49}}, {"head": {"text": "visual inspection", "start": 127, "end": 144}, "tail": {"text": "offset adaptation -LRB- OA -RRB- model", "start": 212, "end": 250}}, {"head": {"text": "automatic speech recognition -LRB- ASR -RRB-", "start": 149, "end": 193}, "tail": {"text": "offset adaptation -LRB- OA -RRB- model", "start": 212, "end": 250}}], "conjunction": [{"head": {"text": "visual inspection", "start": 127, "end": 144}, "tail": {"text": "automatic speech recognition -LRB- ASR -RRB-", "start": 149, "end": 193}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "OA improved phase locking in the auditory nerve -LRB- AN -RRB- and raised ASR accuracy for features derived from AN fibers -LRB- ANFs -RRB- .", "output": {"relations": {"used for": [{"head": {"text": "OA", "start": 0, "end": 2}, "tail": {"text": "phase locking in the auditory nerve -LRB- AN -RRB-", "start": 12, "end": 62}}, {"head": {"text": "OA", "start": 0, "end": 2}, "tail": {"text": "features", "start": 91, "end": 99}}, {"head": {"text": "AN fibers -LRB- ANFs -RRB-", "start": 113, "end": 139}, "tail": {"text": "features", "start": 91, "end": 99}}], "evaluate for": [{"head": {"text": "ASR accuracy", "start": 74, "end": 86}, "tail": {"text": "features", "start": 91, "end": 99}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We also found that OA is crucial for auditory processing by onset neurons -LRB- ONs -RRB- in the next neuronal stage , the auditory brainstem .", "output": {"relations": {"used for": [{"head": {"text": "OA", "start": 19, "end": 21}, "tail": {"text": "auditory processing", "start": 37, "end": 56}}, {"head": {"text": "onset neurons -LRB- ONs -RRB-", "start": 60, "end": 89}, "tail": {"text": "OA", "start": 19, "end": 21}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Multi-layer perceptrons -LRB- MLPs -RRB- performed much better than standard Gaussian mixture models -LRB- GMMs -RRB- for both our ANF-based and ON-based auditory features .", "output": {"relations": {"compare": [{"head": {"text": "Multi-layer perceptrons -LRB- MLPs -RRB-", "start": 0, "end": 40}, "tail": {"text": "Gaussian mixture models -LRB- GMMs -RRB-", "start": 77, "end": 117}}], "used for": [{"head": {"text": "Multi-layer perceptrons -LRB- MLPs -RRB-", "start": 0, "end": 40}, "tail": {"text": "ANF-based and ON-based auditory features", "start": 131, "end": 171}}, {"head": {"text": "Gaussian mixture models -LRB- GMMs -RRB-", "start": 77, "end": 117}, "tail": {"text": "ANF-based and ON-based auditory features", "start": 131, "end": 171}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Recent progress in computer vision has been driven by high-capacity models trained on large datasets .", "output": {"relations": {"used for": [{"head": {"text": "high-capacity models", "start": 54, "end": 74}, "tail": {"text": "computer vision", "start": 19, "end": 34}}, {"head": {"text": "large datasets", "start": 86, "end": 100}, "tail": {"text": "high-capacity models", "start": 54, "end": 74}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Unfortunately , creating large datasets with pixel-level labels has been extremely costly due to the amount of human effort required .", "output": {"relations": {"feature of": [{"head": {"text": "pixel-level labels", "start": 45, "end": 63}, "tail": {"text": "large datasets", "start": 25, "end": 39}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , we present an approach to rapidly creating pixel-accurate semantic label maps for images extracted from modern computer games .", "output": {"relations": {"used for": [{"head": {"text": "approach", "start": 30, "end": 38}, "tail": {"text": "pixel-accurate semantic label maps", "start": 59, "end": 93}}, {"head": {"text": "pixel-accurate semantic label maps", "start": 59, "end": 93}, "tail": {"text": "images", "start": 98, "end": 104}}], "part of": [{"head": {"text": "images", "start": 98, "end": 104}, "tail": {"text": "modern computer games", "start": 120, "end": 141}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We propose a novel step toward the unsupervised seg-mentation of whole objects by combining '' hints '' of partial scene segmentation offered by multiple soft , binary mattes .", "output": {"relations": {"used for": [{"head": {"text": "partial scene segmentation", "start": 107, "end": 133}, "tail": {"text": "unsupervised seg-mentation of whole objects", "start": 35, "end": 78}}, {"head": {"text": "soft , binary mattes", "start": 154, "end": 174}, "tail": {"text": "partial scene segmentation", "start": 107, "end": 133}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "These mattes are implied by a set of hypothesized object boundary fragments in the scene .", "output": {"relations": {"used for": [{"head": {"text": "hypothesized object boundary fragments", "start": 37, "end": 75}, "tail": {"text": "mattes", "start": 6, "end": 12}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This reflects contemporary methods for unsupervised object discovery from groups of images , and it allows us to define intuitive evaluation met-rics for our sets of segmentations based on the accurate and parsimonious delineation of scene objects .", "output": {"relations": {"used for": [{"head": {"text": "contemporary methods", "start": 14, "end": 34}, "tail": {"text": "unsupervised object discovery", "start": 39, "end": 68}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Our proposed approach builds on recent advances in spectral clustering , image matting , and boundary detection .", "output": {"relations": {"used for": [{"head": {"text": "spectral clustering", "start": 51, "end": 70}, "tail": {"text": "approach", "start": 13, "end": 21}}, {"head": {"text": "image matting", "start": 73, "end": 86}, "tail": {"text": "approach", "start": 13, "end": 21}}, {"head": {"text": "boundary detection", "start": 93, "end": 111}, "tail": {"text": "approach", "start": 13, "end": 21}}], "conjunction": [{"head": {"text": "spectral clustering", "start": 51, "end": 70}, "tail": {"text": "image matting", "start": 73, "end": 86}}, {"head": {"text": "image matting", "start": 73, "end": 86}, "tail": {"text": "boundary detection", "start": 93, "end": 111}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "It is demonstrated qualitatively and quantitatively on a dataset of scenes and is suitable for current work in unsupervised object discovery without top-down knowledge .", "output": {"relations": {"used for": [{"head": {"text": "It", "start": 0, "end": 2}, "tail": {"text": "unsupervised object discovery", "start": 111, "end": 140}}], "evaluate for": [{"head": {"text": "dataset of scenes", "start": 57, "end": 74}, "tail": {"text": "It", "start": 0, "end": 2}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Language resource quality is crucial in NLP .", "output": {"relations": {"feature of": [{"head": {"text": "Language resource quality", "start": 0, "end": 25}, "tail": {"text": "NLP", "start": 40, "end": 43}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Many of the resources used are derived from data created by human beings out of an NLP context , especially regarding MT and reference translations .", "output": {"relations": {"hyponym of": [{"head": {"text": "MT", "start": 118, "end": 120}, "tail": {"text": "NLP", "start": 83, "end": 86}}, {"head": {"text": "reference translations", "start": 125, "end": 147}, "tail": {"text": "NLP", "start": 83, "end": 86}}], "conjunction": [{"head": {"text": "MT", "start": 118, "end": 120}, "tail": {"text": "reference translations", "start": 125, "end": 147}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Indeed , automatic evaluations need high-quality data that allow the comparison of both automatic and human translations .", "output": {"relations": {"evaluate for": [{"head": {"text": "high-quality data", "start": 36, "end": 53}, "tail": {"text": "automatic evaluations", "start": 9, "end": 30}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper describes the impact of using different-quality references on evaluation .", "output": {"relations": {"used for": [{"head": {"text": "different-quality references", "start": 41, "end": 69}, "tail": {"text": "evaluation", "start": 73, "end": 83}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Thus , the limitations of the automatic metrics used within MT are also discussed in this regard .", "output": {"relations": {"evaluate for": [{"head": {"text": "automatic metrics", "start": 30, "end": 47}, "tail": {"text": "MT", "start": 60, "end": 62}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This poster paper describes a full scale two-level morphological description -LRB- Karttunen , 1983 ; Koskenniemi , 1983 -RRB- of Turkish word structures .", "output": {"relations": {"used for": [{"head": {"text": "full scale two-level morphological description", "start": 30, "end": 76}, "tail": {"text": "Turkish word structures", "start": 130, "end": 153}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The description has been implemented using the PC-KIMMO environment -LRB- Antworth , 1990 -RRB- and is based on a root word lexicon of about 23,000 roots words .", "output": {"relations": {"used for": [{"head": {"text": "PC-KIMMO environment", "start": 47, "end": 67}, "tail": {"text": "description", "start": 4, "end": 15}}, {"head": {"text": "root word lexicon", "start": 114, "end": 131}, "tail": {"text": "description", "start": 4, "end": 15}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Turkish is an agglutinative language with word structures formed by productive affixations of derivational and inflectional suffixes to root words .", "output": {"relations": {"hyponym of": [{"head": {"text": "Turkish", "start": 0, "end": 7}, "tail": {"text": "agglutinative language", "start": 14, "end": 36}}], "feature of": [{"head": {"text": "word structures", "start": 42, "end": 57}, "tail": {"text": "agglutinative language", "start": 14, "end": 36}}], "part of": [{"head": {"text": "productive affixations of derivational and inflectional suffixes", "start": 68, "end": 132}, "tail": {"text": "word structures", "start": 42, "end": 57}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The surface realizations of morphological constructions are constrained and modified by a number of phonetic rules such as vowel harmony .", "output": {"relations": {"used for": [{"head": {"text": "phonetic rules", "start": 100, "end": 114}, "tail": {"text": "surface realizations of morphological constructions", "start": 4, "end": 55}}], "hyponym of": [{"head": {"text": "vowel harmony", "start": 123, "end": 136}, "tail": {"text": "phonetic rules", "start": 100, "end": 114}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper deals with the problem of generating the fundamental frequency -LRB- F0 -RRB- contour of speech from a text input for text-to-speech synthesis .", "output": {"relations": {"used for": [{"head": {"text": "fundamental frequency -LRB- F0 -RRB- contour of speech", "start": 52, "end": 106}, "tail": {"text": "text-to-speech synthesis", "start": 129, "end": 153}}, {"head": {"text": "text input", "start": 114, "end": 124}, "tail": {"text": "fundamental frequency -LRB- F0 -RRB- contour of speech", "start": 52, "end": 106}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We have previously introduced a statistical model describing the generating process of speech F0 contours , based on the discrete-time version of the Fujisaki model .", "output": {"relations": {"used for": [{"head": {"text": "statistical model", "start": 32, "end": 49}, "tail": {"text": "speech F0 contours", "start": 87, "end": 105}}, {"head": {"text": "Fujisaki model", "start": 150, "end": 164}, "tail": {"text": "statistical model", "start": 32, "end": 49}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "One remarkable feature of this model is that it has allowed us to derive an efficient algorithm based on powerful statistical methods for estimating the Fujisaki-model parameters from raw F0 contours .", "output": {"relations": {"feature of": [{"head": {"text": "remarkable feature", "start": 4, "end": 22}, "tail": {"text": "model", "start": 31, "end": 36}}], "used for": [{"head": {"text": "algorithm", "start": 86, "end": 95}, "tail": {"text": "Fujisaki-model parameters", "start": 153, "end": 178}}, {"head": {"text": "statistical methods", "start": 114, "end": 133}, "tail": {"text": "algorithm", "start": 86, "end": 95}}, {"head": {"text": "raw F0 contours", "start": 184, "end": 199}, "tail": {"text": "Fujisaki-model parameters", "start": 153, "end": 178}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "To associate a sequence of the Fujisaki-model parameters with a text input based on statistical learning , this paper proposes extending this model to a context-dependent one .", "output": {"relations": {"used for": [{"head": {"text": "text input", "start": 64, "end": 74}, "tail": {"text": "Fujisaki-model parameters", "start": 31, "end": 56}}, {"head": {"text": "statistical learning", "start": 84, "end": 104}, "tail": {"text": "Fujisaki-model parameters", "start": 31, "end": 56}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We further propose a parameter training algorithm for the present model based on a decision tree-based context clustering .", "output": {"relations": {"used for": [{"head": {"text": "parameter training algorithm", "start": 21, "end": 49}, "tail": {"text": "model", "start": 66, "end": 71}}, {"head": {"text": "decision tree-based context clustering", "start": 83, "end": 121}, "tail": {"text": "parameter training algorithm", "start": 21, "end": 49}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We introduce a method to accelerate the evaluation of object detection cascades with the help of a divide-and-conquer procedure in the space of candidate regions .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 15, "end": 21}, "tail": {"text": "evaluation of object detection cascades", "start": 40, "end": 79}}, {"head": {"text": "divide-and-conquer procedure", "start": 99, "end": 127}, "tail": {"text": "method", "start": 15, "end": 21}}], "feature of": [{"head": {"text": "space of candidate regions", "start": 135, "end": 161}, "tail": {"text": "divide-and-conquer procedure", "start": 99, "end": 127}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Compared to the exhaustive procedure that thus far is the state-of-the-art for cascade evaluation , the proposed method requires fewer evaluations of the classifier functions , thereby speeding up the search .", "output": {"relations": {"used for": [{"head": {"text": "exhaustive procedure", "start": 16, "end": 36}, "tail": {"text": "cascade evaluation", "start": 79, "end": 97}}, {"head": {"text": "method", "start": 113, "end": 119}, "tail": {"text": "search", "start": 201, "end": 207}}], "compare": [{"head": {"text": "exhaustive procedure", "start": 16, "end": 36}, "tail": {"text": "method", "start": 113, "end": 119}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Furthermore , we show how the recently developed efficient subwindow search -LRB- ESS -RRB- procedure -LSB- 11 -RSB- can be integrated into the last stage of our method .", "output": {"relations": {"part of": [{"head": {"text": "subwindow search -LRB- ESS -RRB- procedure", "start": 59, "end": 101}, "tail": {"text": "method", "start": 162, "end": 168}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This allows us to use our method to act not only as a faster procedure for cascade evaluation , but also as a tool to perform efficient branch-and-bound object detection with nonlinear quality functions , in particular kernel-ized support vector machines .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 26, "end": 32}, "tail": {"text": "cascade evaluation", "start": 75, "end": 93}}, {"head": {"text": "method", "start": 26, "end": 32}, "tail": {"text": "branch-and-bound object detection", "start": 136, "end": 169}}, {"head": {"text": "nonlinear quality functions", "start": 175, "end": 202}, "tail": {"text": "branch-and-bound object detection", "start": 136, "end": 169}}], "hyponym of": [{"head": {"text": "kernel-ized support vector machines", "start": 219, "end": 254}, "tail": {"text": "nonlinear quality functions", "start": 175, "end": 202}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Experiments on the PASCAL VOC 2006 dataset show an acceleration of more than 50 % by our method compared to standard cascade evaluation .", "output": {"relations": {"evaluate for": [{"head": {"text": "PASCAL VOC 2006 dataset", "start": 19, "end": 42}, "tail": {"text": "method", "start": 89, "end": 95}}, {"head": {"text": "PASCAL VOC 2006 dataset", "start": 19, "end": 42}, "tail": {"text": "cascade evaluation", "start": 117, "end": 135}}], "compare": [{"head": {"text": "cascade evaluation", "start": 117, "end": 135}, "tail": {"text": "method", "start": 89, "end": 95}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Background modeling is an important component of many vision systems .", "output": {"relations": {"part of": [{"head": {"text": "Background modeling", "start": 0, "end": 19}, "tail": {"text": "vision systems", "start": 54, "end": 68}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "When the scene exhibits a persistent dynamic behavior in time , such an assumption is violated and detection performance deteriorates .", "output": {"relations": {"feature of": [{"head": {"text": "persistent dynamic behavior", "start": 26, "end": 53}, "tail": {"text": "scene", "start": 9, "end": 14}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , we propose a new method for the modeling and subtraction of such scenes .", "output": {"relations": {"used for": [{"head": {"text": "method", "start": 33, "end": 39}, "tail": {"text": "modeling and subtraction of such scenes", "start": 48, "end": 87}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Towards the modeling of the dynamic characteristics , optical flow is computed and utilized as a feature in a higher dimensional space .", "output": {"relations": {"used for": [{"head": {"text": "optical flow", "start": 54, "end": 66}, "tail": {"text": "modeling of the dynamic characteristics", "start": 12, "end": 51}}, {"head": {"text": "optical flow", "start": 54, "end": 66}, "tail": {"text": "feature", "start": 97, "end": 104}}, {"head": {"text": "feature", "start": 97, "end": 104}, "tail": {"text": "modeling of the dynamic characteristics", "start": 12, "end": 51}}], "feature of": [{"head": {"text": "higher dimensional space", "start": 110, "end": 134}, "tail": {"text": "feature", "start": 97, "end": 104}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Inherent ambiguities in the computation of features are addressed by using a data-dependent bandwidth for density estimation using kernels .", "output": {"relations": {"feature of": [{"head": {"text": "ambiguities", "start": 9, "end": 20}, "tail": {"text": "computation of features", "start": 28, "end": 51}}], "used for": [{"head": {"text": "data-dependent bandwidth", "start": 77, "end": 101}, "tail": {"text": "ambiguities", "start": 9, "end": 20}}, {"head": {"text": "data-dependent bandwidth", "start": 77, "end": 101}, "tail": {"text": "density estimation", "start": 106, "end": 124}}, {"head": {"text": "kernels", "start": 131, "end": 138}, "tail": {"text": "density estimation", "start": 106, "end": 124}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "In this paper , we present our approach for using information extraction annotations to augment document retrieval for distillation .", "output": {"relations": {"used for": [{"head": {"text": "information extraction annotations", "start": 50, "end": 84}, "tail": {"text": "document retrieval for distillation", "start": 96, "end": 131}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "This paper presents a novel representation for three-dimensional objects in terms of affine-invariant image patches and their spatial relationships .", "output": {"relations": {"used for": [{"head": {"text": "representation", "start": 28, "end": 42}, "tail": {"text": "three-dimensional objects", "start": 47, "end": 72}}], "feature of": [{"head": {"text": "affine-invariant image patches", "start": 85, "end": 115}, "tail": {"text": "three-dimensional objects", "start": 47, "end": 72}}, {"head": {"text": "spatial relationships", "start": 126, "end": 147}, "tail": {"text": "affine-invariant image patches", "start": 85, "end": 115}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Multi-view constraints associated with groups of patches are combined with a normalized representation of their appearance to guide matching and reconstruction , allowing the acquisition of true three-dimensional affine and Euclidean models from multiple images and their recognition in a single photograph taken from an arbitrary viewpoint .", "output": {"relations": {"conjunction": [{"head": {"text": "Multi-view constraints", "start": 0, "end": 22}, "tail": {"text": "normalized representation", "start": 77, "end": 102}}, {"head": {"text": "matching", "start": 132, "end": 140}, "tail": {"text": "reconstruction", "start": 145, "end": 159}}], "used for": [{"head": {"text": "Multi-view constraints", "start": 0, "end": 22}, "tail": {"text": "matching", "start": 132, "end": 140}}, {"head": {"text": "Multi-view constraints", "start": 0, "end": 22}, "tail": {"text": "reconstruction", "start": 145, "end": 159}}, {"head": {"text": "normalized representation", "start": 77, "end": 102}, "tail": {"text": "matching", "start": 132, "end": 140}}, {"head": {"text": "normalized representation", "start": 77, "end": 102}, "tail": {"text": "reconstruction", "start": 145, "end": 159}}, {"head": {"text": "images", "start": 255, "end": 261}, "tail": {"text": "acquisition of true three-dimensional affine and Euclidean models", "start": 175, "end": 240}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "The proposed approach does not require a separate segmentation stage and is applicable to cluttered scenes .", "output": {"relations": {"used for": [{"head": {"text": "approach", "start": 13, "end": 21}, "tail": {"text": "cluttered scenes", "start": 90, "end": 106}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Fast algorithms for nearest neighbor -LRB- NN -RRB- search have in large part focused on 2 distance .", "output": {"relations": {"used for": [{"head": {"text": "Fast algorithms", "start": 0, "end": 15}, "tail": {"text": "nearest neighbor -LRB- NN -RRB- search", "start": 20, "end": 58}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "Here we develop an approach for 1 distance that begins with an explicit and exactly distance-preserving embedding of the points into 2 2 .", "output": {"relations": {"used for": [{"head": {"text": "approach", "start": 19, "end": 27}, "tail": {"text": "1 distance", "start": 32, "end": 42}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We show how this can efficiently be combined with random-projection based methods for 2 NN search , such as locality-sensitive hashing -LRB- LSH -RRB- or random projection trees .", "output": {"relations": {"conjunction": [{"head": {"text": "this", "start": 12, "end": 16}, "tail": {"text": "random-projection based methods", "start": 50, "end": 81}}, {"head": {"text": "locality-sensitive hashing -LRB- LSH -RRB-", "start": 108, "end": 150}, "tail": {"text": "random projection trees", "start": 154, "end": 177}}], "used for": [{"head": {"text": "random-projection based methods", "start": 50, "end": 81}, "tail": {"text": "NN search", "start": 88, "end": 97}}], "hyponym of": [{"head": {"text": "locality-sensitive hashing -LRB- LSH -RRB-", "start": 108, "end": 150}, "tail": {"text": "random-projection based methods", "start": 50, "end": 81}}, {"head": {"text": "random projection trees", "start": 154, "end": 177}, "tail": {"text": "random-projection based methods", "start": 50, "end": 81}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}
{"input": "We rigorously establish the correctness of the methodology and show by experimentation using LSH that it is competitive in practice with available alternatives .", "output": {"relations": {"compare": [{"head": {"text": "it", "start": 102, "end": 104}, "tail": {"text": "alternatives", "start": 147, "end": 159}}]}}, "schema": {"relations": [{"label": "used for", "description": "B is used for A, B models A, A is trained on B, B exploits A, A is based on B."}, {"label": "evaluate for", "description": "B evaluates A, B is a metric for A, A is evaluated on/by B."}, {"label": "feature of", "description": "B belongs to A, B is a feature of A, B is under A domain."}, {"label": "hyponym of", "description": "B is a hyponym of A, B is a type of A."}, {"label": "part of", "description": "B is a part of A, A includes B, A contains B."}, {"label": "compare", "description": "Symmetric relation (use blue to denote entity). Opposite of conjunction, compare two models/methods, or listing two opposing entities."}, {"label": "conjunction", "description": "Symmetric relation (use blue to denote entity). Function as similar role or use/incorporate with."}]}}