Code and Data for our ACL paper "Men Are Elected, Women Are Married: Events Gender Bias on Wikipedia"
Please cite us if you find any of these content useful.
@inproceedings{sun2021men,
title = {Men Are Elected, Women Are Married: Events Gender Bias on Wikipedia},
author = {Sun, Jiao and Peng, Nanyun},
booktitle = {Proceedings of the Conference of the 59th Annual Meeting of the Association for Computational Linguistics (ACL)},
year = {2021}
}
- Python>=3.7.6
- Analysis Pipeline. We demostrate our analysis framework for Personal Life section in
personal-life-reproduce.ipynb
, where you can find the code for key steps that we are using in the paper, including Acquire templates by name swaping, odds ratio calculation, replicate figures as Figure1, and etc . - WEAT calculation. The code for replicating our WEAT calculation is under
WEAT.py
, please see how to use in./weat.sh
to replicate the result. You can simply change the list of dictionary inartists.json
and use our script to measure the associated gender bias of any two lists of attributes.
-
Please find all collected data under here where you can access all sections that we collected, statistics in Appendix Table 6, and put them under
data/raw-data
-
The data we use fo analyzing the career and Personal Life sections are under
data/ee-fm
-
The intermediate results which contain extracted events for each section are under
data/ee-model
-
Please download the embeddings under here and put them under
WEAT/embeddings/
folder