Skip to content

A light-weighted UMLS-based data augmentation for biomedical NLP tasks including Named Entity Recognition and sentence classification.

License

Notifications You must be signed in to change notification settings

WengLab-InformaticsResearch/UMLS-EDA

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

UMLS-EDA :octocat:

🎉 A light-weighted UMLS-based data augmentation for biomedical NLP tasks including Named Entity Recognition and sentence classification 🎉

User Guide

0. Before start

  1. Install 'UMLS' and 'QuickUMLS' locally
  2. Get your UMLS SOAP API Key from the UTS ‘My Profile’ area after signing in UMLS Terminology service
  3. Add your API Key and QuickUMLS directory to the config.py.
  4. Costomzie other variables in the config.py

1. Named Entity Recognition

  • Input: CoNLL format file
  • Usage:
    python augment4ner.py [-h] --input INPUT [--output OUTPUT] [--num_aug NUM_AUG] [--alpha ALPHA]

2. Sentence Classification

  • Input: "|" seperated file (index|label|sentence text)
  • Usage:
    python augment4class.py [-h] --input INPUT [--output OUTPUT] [--num_aug NUM_AUG] [--alpha ALPHA]

See examples/example4ner.conll and example/example4class.txt

Reference

About

A light-weighted UMLS-based data augmentation for biomedical NLP tasks including Named Entity Recognition and sentence classification.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages