BioCreative II Gene Mention Tokenized Training and Validation Data

Giving access to tokenized training and validation data for the BioCreative II gene mention task. It includes sequences of tokens, case-folded tokens, part-of-speech tags, and BIO sequence labels.

Use Cases

• Train a named entity recognition model to identify gene mentions using the BIO sequence tags (labels) on token sequences.
• Build a model that leverages part-of-speech tags (tags) to improve the accuracy of gene mention extraction from biomedical text.
• Compare the performance of models using standard tokens versus case-folded tokens (folded_tokens) for gene mention detection.
• Develop a sequence labeling pipeline for biomedical text using the provided token and label sequences.

Strengths

• Data is specifically formatted for the established BioCreative II gene mention task.
• Includes multiple linguistic features per sequence: tokens, case-folded variants, part-of-speech tags, and BIO labels.

Limitations

• The dataset size, row count, and column structure are not specified.
• The data's temporal and geographic scope is unknown, limiting analysis of trends or regional applicability.

Provenance

Source

huggingface, uploaded by author 'enoriega'.

Collection Method

Data tokenized with processors from clulab/processors GitHub repository for the BioCreative II task.

Freshness

Last updated on 2022-04-20.