Home

Awesome

FAHT

The implementation of FAHT (IJCAI 19), a fair classifier for online stream based decision-making. Detailded information about FAHT is provided in FAHT.

Instructions

  1. Clone this repository.
  2. Download the datasets as described in the Experiment/Data folder of this repository to the root folder of the project.
  3. Run the code with Weka > 3.9.
    *In Experiment folder: InstanceStreamClassifier.java and WindowStreamClassifier.java evaluate the landmark window model and sliding window model, respectively.
    *The FAHT folder contains the source code of the proposed FAHT.

Citation

@inproceedings{zhang2019faht,
title={FAHT: an adaptive fairness-aware decision tree classifier},
author={Zhang, Wenbin and Ntoutsi, Eirini},
booktitle={Proceedings of the 28th International Joint Conference on Artificial Intelligence},
pages={1480--1486},
year={2019},
organization={AAAI Press}
}