Weakly Supervised Sequence Tagging from Noisy Rules - Reproducibility Code

Getting Started

These instructions will get you a copy of our experiments up and running on your local machine for development and testing purposes.

Installing

In your virtual environment, please install the required dependencies using

pip install -r requirements.txt

Or alternatively

conda install --file requirements.txt

Datasets

Our experiments depend on six different datasets that you will need to download separately.

• BC5CDR: Download and install the train, development, and test BioCreative V CDR corpus data files. Place the three separate files inside data/BC5CD

• NCBI Disease: Download and install the complete training, development, and testing sets. Place the three separate files inside data/NCBI.

• LaptopReview: Download the train data V2.0 for the Laptops and Restaurants dataset, and place the Laptop_Train_v2.xml file inside data/LaptopReview. Then, download the test data - phase B, and place the Laptops_Test_Data_phaseB.xml file inside the same directory.

• CoNLL v5: Download and compile the English dataset version 5.0, and place it in data/conll-formatted-ontonotes-5.0.

• Scibert: Download the scibert-scivocab-uncased version of the Scibert embeddings, and place the files *weights.tar.gz and vocab.txt inside data/scibert_scibocab_uncased.

• UMLS: The UMLS dictionaries have been extracted from the UMLS 2018AB dataset and are provided in our code. They are distributed according to the License Agreement for Use of the UMLS® Metathesaurus®.

• AutoNER Dictionaries. The AutoNER dictionaries for the BC5CDR, LaptopReview, and NCBI datasets have been generously provided by Jingbo Shang et al. They have been sourced from the EMNLP 2018 paper "Learning Named Entity Tagger using Domain-Specific Dictionary".

Citation

Please cite the following paper if you are using our tool. Thank you!

Safranchik Esteban, Shiying Luo, Stephen H. Bach. "Weakly Supervised Sequence Tagging From Noisy Rules". In 34th AAAI Conference on Artificial Intelligence, 2020.

@inproceedings{safranchik2020weakly,
  title = {Weakly Supervised Sequence Tagging From Noisy Rules}, 
  author = {Safranchik, Esteban and Luo, Shiying and Bach, Stephen H.}, 
  booktitle = {AAAI}, 
  year = 2020, 
}