candle-fire / data /tools /extract_entities.json
KevinIsInCoding
feat: initial project scaffold and Stage 2 ingestion pipeline
0ccfe4a
Raw
History Blame Contribute Delete
2.25 kB
{
"type": "object",
"properties": {
"pmid": {
"type": "string",
"description": "The PubMed ID of the paper being analyzed."
},
"entities": {
"type": "array",
"description": "Biomedical entities identified in the abstract.",
"items": {
"type": "object",
"properties": {
"type": {
"type": "string",
"enum": ["Gene", "Protein", "Compound", "Pathway", "Phenotype", "Mechanism"],
"description": "Category of the biomedical entity."
},
"name": {
"type": "string",
"description": "Exact name as it appears in the text."
},
"confidence": {
"type": "number",
"minimum": 0.0,
"maximum": 1.0,
"description": "Confidence that this is correctly identified (1.0 = unambiguous)."
},
"mentions": {
"type": "integer",
"minimum": 1,
"description": "Number of times this entity is mentioned in the abstract."
}
},
"required": ["type", "name", "confidence", "mentions"]
}
},
"relationships": {
"type": "array",
"description": "Relationships between entities identified in the abstract.",
"items": {
"type": "object",
"properties": {
"source": {
"type": "string",
"description": "Name of the source entity (must match an entity in the entities list)."
},
"target": {
"type": "string",
"description": "Name of the target entity (must match an entity in the entities list)."
},
"type": {
"type": "string",
"enum": ["BINDS", "INHIBITS", "ASSOCIATED_WITH", "TESTED_IN", "EXPRESSED_IN", "CO_OCCURS"],
"description": "Type of relationship between source and target."
},
"evidence_text": {
"type": "string",
"description": "Brief excerpt from the abstract that supports this relationship."
}
},
"required": ["source", "target", "type", "evidence_text"]
}
}
},
"required": ["pmid", "entities", "relationships"]
}