FairRegBoost: An End-to-End Data Processing Framework for Fair and Scalable Regression

This repository contains the codes needed to reproduce the experiments of our submitted CIKM 2025 Paper

General Information

For the experiments we reuse work from other GitHubs (put in subfolders of algorithms)
For some approaches, some slight adaptations had to be made to integrate them in our framework, but the general approach was not altered.
The experiments were run under MacOS Sonoma 14.4, Python Version 3.9.6.

You need to run run_exps_reg.py. The parameters to set for the experiments, like datasets, models to train, are in that file.
This will call the main.py function that runs the experiments, and subsequently calls evaluation_reg.py to evaluate the overall results.
An evaluation file is automatically generated for an experiment, which shows metrics for RMSE and W2, along additional metrics.

Name		Name	Last commit message	Last commit date
Latest commit History 2 Commits
Datasets		Datasets
algorithm		algorithm
configs		configs
README.md		README.md
evaluation_reg.py		evaluation_reg.py
main.py		main.py
requirements.txt		requirements.txt
run_exps_reg.py		run_exps_reg.py