hogunpark / EJGAN

Incorporating Experts' Judgment into Machine Learning Models

This repository provides a reference implementation of EJ-GAN as described in the paper:

Incorporating Experts' Judgment into Machine Learning Models
Hogun Park, Aly Megahed, Peifeng Yin, Yuya Ong, Pravar Mahajan, Pei Guo
Expert Systems with Applications, Elsevier, 2023.

We note that our code was tested under Python 3.6.10.

Usage for running GAN-based Conflict Resolution Function

Example

To run EJ-GAN on CREDIT dataset, execute the following command from the project home directory:

usage: main_GAN_Classification.py [-h] --input INPUT
                              [--export_plot EXPORT_PLOT] [--prior PRIOR]
                              [--weight WEIGHT] [--num-steps NUM_STEPS]
                              [--hidden-size HIDDEN_SIZE]
                              [--dc_weight DC_WEIGHT]
                              [--batch-size BATCH_SIZE]
                              [--log-every LOG_EVERY] [--n_bins N_BINS]
                              [--classifier CLASSIFIER]


		e.g., --input "data/credit/input.csv" --prior 4 --classifier 0  # 4 was used for prior in our experiment

Citing

If you find EJ-GAN useful for your research, please consider citing the following paper:

@article{ejgan-eswa,
 author = {Park, Hogun and Megahed, Aly and Yin, Peifeng and Ong, Yuya and Mahajan, Pravar and Guo, Pei},
 title = {Incorporating Experts' Judgment into Machine Learning Models},
 journal ={Expert Systems with Applications},
 year = {2023},
 publisher={Elsevier}
}

ACKNOWLEDGEMENT

The implementation of this repository is porked and based on the codebase.
@Copyright 2020 IBM Corporation
Licensed under the Apache License, Version 2.0 (the "License").
You may not use this file except in compliance with the License.