AI Gender-bias paraphrasing

GPT3 used to remove bias from the job adverts

Medium article: https://medium.com/axel-springer-tech/using-ai-to-tackle-gender-bias-in-job-adverts-a69958953e91

Overview

This is a demo of the application for removal of the gender-biased term from job adverts. A static website ready to deploy. This repo includes all necessary file to deply it to i.e Heroku.

• .env
• .gitignore
• app
• Procfile
• README.md
• requirements.txt
• runtime.txt
• static
• templates

Installation & Usage

Remeber to set env variables with OPEN_AI_API_KEY. (see the top of the backend)

# change the directory
$ cd fastapi-web-starter
# install packages
$ pip install -r requirements.txt
# start the server
$ uvicorn app.main:app --reload --port 8080

Visit http://127.0.0.1:8080/.

Features

• Highlighting of gender-biased term
• Automatic paraphrasing

Test

All tests are under tests directory.

# Change the directory
$ cd fastapi-web-starter
# Run tests
$ pytest -v

Credits

Original paper which helped me build a list of the biased words : https://www.sussex.ac.uk/webteam/gateway/file.php?name=gendered-wording-in-job-adverts.pdf&site=7

Licence

Boilerplate Frontend: Copyright 2013-2020 Start Bootstrap LLC. Code released under the MIT license. Boilerplate Backend : Copyright 2021 Shinichi Okada