---
license: mit
task_categories:
- token-classification
language:
- en
dataset_info:
  features:
  - name: id
    dtype: string
  - name: tokens
    sequence: string
  - name: ner_tags
    sequence:
      class_label:
        names:
          '0': O
          '1': B-DiseaseClass
          '2': I-DiseaseClass
          '3': B-SpecificDisease
          '4': I-SpecificDisease
          '5': B-CompositeMention
          '6': I-CompositeMention
          '7': B-Modifier
          '8': I-Modifier
  splits:
  - name: test
    num_bytes: 319540
    num_examples: 80
  - name: train
    num_bytes: 3086873
    num_examples: 792
  download_size: 854342
  dataset_size: 3406413
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---

## Benchmark dataset NCBI
This dataset was generated by the Data preprocessing step of the NERFAIR workflow (More information: https://github.com/YasCoMa/ner-fair-workflow )

Original dataset: Doğan, Rezarta Islamaj, Robert Leaman, and Zhiyong Lu. 2014. “NCBI Disease Corpus: A Resource for Disease Name Recognition and Concept Normalization.” Journal of Biomedical Informatics 47 (February): 1–10.