FairSIN

Code for paper FairSIN: Achieving Fairness in Graph Neural Networks through Sensitive Information Neutralization https://github.com/Landon5282/FairSIN

To install requirements:

pip install -r requirements.txt

The instructions for the relevant experiments are all saved in experiment.sh.

The reweight.py script contains the implementation of a data-centric variant, while the in-train.py script is for the model-centric variant.