Wet Lab Protocols

This repository contains the Conditional Random Fields (CRF) model for the shared task Entity and relation recognition over wet-lab protocols. This shared task was part of the 6th Workshop on Noisy User-generated Text held in 2020.

Task: Named Entity Recognition

The task involves extracting 18 entity classes from Wet Lab Protocols. The entity classes are described in An Annotated Corpus for Machine Reading of Instructions in Wet Lab Protocols by Kulkarni et al(2018). Have a look at Annotation Guidelines section of the paper for detailed description of the entity classes.

Example:

Sentence #6 shows the following entities:

Entity	Text
Action	Mix
Modifier	thoroughly
Action	pulse-spin
Device	microfuge

BRAT styled annotated protocols

http://kb1.cse.ohio-state.edu:8010/index.xhtml#/wnut_20_data/

Entity, Action and Relation extraction demo

• http://bionlp.osu.edu:5000/protocols
• Demo page of the system developed as part of Kulkarni et al's 2018 paper.

Data

• https://github.com/jeniyat/WNUT_2020_NER
• https://github.com/jeniyat/WNUT_2020_RE

How to run?

Below are the example commands.

• Load protocol and entity annotations(if available)

  • Segments protocol into sentences and words.
  • Executes spaCy's NLP pipeline over the protocol.

    python -m src.dataset --ann_format standoff --protocol_id 101

• Execute Conditional Random Fields (CRF) model

  • Train CRF model

      python -m src.crf

    • Saves model in ./output/models directory

  • Validate development set

      python -m src.crf --train_model ./output/models/model_standoff.pkl --evaluate_collection

  • Predict on test set

      python -m src.crf --train_model ./output/models/model_standoff.pkl --predict_collection

Publication

KaushikAcharya at WNUT 2020 Shared Task-1: Conditional Random Field(CRF) based Named Entity Recognition(NER) for Wet Lab Protocols (EMNLP | WNUT)

Citation

If you find this implementation helpful, please consider citing:

@inproceedings{acharya-2020-wnut,
    title = "{WNUT} 2020 Shared Task-1: Conditional Random Field({CRF}) based Named Entity Recognition({NER}) for Wet Lab Protocols",
    author = "Acharya, Kaushik",
    booktitle = "Proceedings of the Sixth Workshop on Noisy User-generated Text (W-NUT 2020)",
    month = nov,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://www.aclweb.org/anthology/2020.wnut-1.37",
    pages = "286--289",
    abstract = "The paper describes how classifier model built using Conditional Random Field detects named entities in wet lab protocols.",
}

Related Work

• MEDDOPROF
  • https://github.com/kaushikacharya/clinical_occupation_recognition
  • Named Entity Recognition and Normalization on Spanish medical documents.