fair-binary-classification

Fairness-aware binary classification for the Adult Census Income dataset

Status: Personal / university coursework archive — not actively maintained. Dependencies and tooling may be outdated.

How to run

Use Python 3 with Jupyter. Install pandas, scikit-learn, and aif360 (IBM AI Fairness 360) plus their dependencies, then open adult_basic.ipynb and adult_fair.ipynb and run sequentially.

Contents

This repository contains two jupyter notebooks for predictive modeling of the adult dataset:

• adult_basic.ipynb includes a fairness-unaware approach using pandas and scikit-learn
• adult_fair.ipynb utilizes AIF360 toolset and its metrics and bias mitigation techniques to reduce bias w.r.t. sensitive attributes present in the dataset

Both notebooks provide detailed information about the analysis and the algorithms that are utilized within.