bltlab/open-ner-core-types

Dataset Card for OpenNER 1.0

OpenNER 1.0 is a standardized collection of openly-available named entity recognition (NER) datasets. OpenNER contains 36 NER corpora that span 52 languages, human-annotated in varying named entity ontologies. We correct annotation format issues, standardize the original datasets into a uniform representation with consistent entity type names across corpora, and provide the collection in a structure that enables research in multilingual and multi-ontology NER.

This version of the dataset contains only the core entity types of Person (PER), Location (LOC), and Organization (ORG).

Dataset Details

Dataset Description

• Curated by: BLT Lab: Chester Palen-Michel, Maxwell Pickering, Maya Kruse, Jonne Sälevä, & Constantine Lignos
• Shared by: Chester Palen-Michel
• Language(s) (NLP): Akan/Twi, Amharic, Arabic, Bambara, Basque, Bavarian German, Catalan, Chichewa, chiShona, Croatian, Danish, Dutch, English, Éwé, Finnish, Fon, Galician, German, Ghomálá', Greek, Hausa, Hebrew, Hindi, Igbo, isiXhosa, Italian, Japanese, Kazakh, Kinyarwanda, Kiswahili, Luganda, Luo, Maghrebi Arabic, Mandarin Chinese, Marathi, Mossi, Naija, Nepali, Norwegian (Bokmål), Norwegian (Nynorsk), Persian Farsi, Portuguese, Romanian, Setswana, Slovak, Slovenian, Spanish, Swedish, Thai, Wolof, Yorùbá, Zulu
• License: CC-BY 4.0 for the OpenNER collection. Individual datasets have their own licenses.

Dataset Sources

• Repository: OpenNER Github
• Paper: ACL Anthology and arXiv

Uses

Primarily to be used for research regarding multilingual NER with different entity sets and annotation guidelines.

Dataset Structure

{
  'id': '0',
  'tokens': ['Melbourne', '(', 'Australia', ')', ',', '25', 'may', '(', 'EFE', ')', '.'],
  'ner_tags': [5, 0, 5, 0, 0, 0, 0, 0, 3, 0, 0]
}

The ner_tags column includes an internal int2str map to access the corresponding string label for each integer label.

Dataset Creation

Curation Rationale

There are many NER datasets but many have different formats, label names, and label schemas. OpenNER standardizes and collects these many NER datasets in easily accessible format and place.

Source Data

OpenNER primarily contains newswire and web text. See our paper with each cited paper describing each dataset for details of source data.

Data Collection and Processing

See our paper for details on data collection and processing.

Who are the source data producers?

See our paper with each cited paper describing each dataset for details of source data.

Annotations

Annotation is named entity recognition annotation on each token in BIO format.

Annotation process

See our paper with each cited paper for each dataset for details on the annotation process.

Who are the annotators?

OpenNER is a collection of many existing datasets which have been human annotated. Annotator details are included in each dataset's original publication. Citations for each work can be found in our paper.

Personal and Sensitive Information

To the best of our knowledge there is no personal or sensitive information beyond that which generally occurs in newswire text.

Bias, Risks, and Limitations

See our paper for discussion of limitations and biases.

Recommendations

See our paper for recommendations regarding OpenNER.

Citation

If you make use of this dataset, please cite our paper using this bibtex:

BibTeX:

@inproceedings{palen-michel-etal-2025-openner,
    title = "{O}pen{NER} 1.0: Standardized Open-Access Named Entity Recognition Datasets in 50+ Languages",
    author = {Palen-Michel, Chester  and
      Pickering, Maxwell  and
      Kruse, Maya  and
      S{\"a}lev{\"a}, Jonne  and
      Lignos, Constantine},
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2025.emnlp-main.1708/",
    doi = "10.18653/v1/2025.emnlp-main.1708",
    pages = "33637--33662",
    ISBN = "979-8-89176-332-6",
    abstract = "We present OpenNER 1.0, a standardized collection of openly-available named entity recognition (NER) datasets.OpenNER contains 36 NER corpora that span 52 languages, human-annotated in varying named entity ontologies.We correct annotation format issues, standardize the original datasets into a uniform representation with consistent entity type names across corpora, and provide the collection in a structure that enables research in multilingual and multi-ontology NER.We provide baseline results using three pretrained multilingual language models and two large language models to compare the performance of recent models and facilitate future research in NER.We find that no single model is best in all languages and that significant work remains to obtain high performance from LLMs on the NER task.OpenNER is released at https://github.com/bltlab/open-ner."
}

Dataset Card Authors

Chester Palen-Michel @cpalenmichel

Dataset Card Contact

Chester Palen-Michel @cpalenmichel