Awesome
Good Secretaries, Bad Truck Drivers? Occupational Gender Stereotypes in Sentiment Analysis
This repository contains the code for the following paper:
Good Secretaries, Bad Truck Drivers? Occupational Gender Stereotypes in Sentiment Analysis
Jayadev Bhaskaran and Isha Bhallamudi
to appear at ACL GeBNLP (Workshop on Gender Bias for NLP) at ACL 2019, Florence, Italy.
The final paper can be found here: https://arxiv.org/abs/1906.1025
Repository Structure
- data/gender_corpus.tsv - New corpus to evaluate occupational gender stereotypes in sentiment analysis models.
- data/ - Directory that contains all data (SST-2, evaluation corpus).
- runs/ - Directory that contains training run outputs (model predictions on evaluation corpus, model checkpoints).
- src/ - Directory for all source code (e.g. models, preprocessing scripts, analysis scripts).
For any questions, please contact jayadevbhaskaran@gmail.com or raise an issue here.