Software Engineering Principles for Fairer Systems: Experiments with GroupCART

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Peng, Kewen, Zhuo, Hao, Yang, Yicheng, Menzies, Tim
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912332640157696
author Peng, Kewen
Zhuo, Hao
Yang, Yicheng
Menzies, Tim
author_facet Peng, Kewen
Zhuo, Hao
Yang, Yicheng
Menzies, Tim
contents Discrimination-aware classification aims to make accurate predictions while satisfying fairness constraints. Traditional decision tree learners typically optimize for information gain in the target attribute alone, which can result in models that unfairly discriminate against protected social groups (e.g., gender, ethnicity). Motivated by these shortcomings, we propose GroupCART, a tree-based ensemble optimizer that avoids bias during model construction by optimizing not only for decreased entropy in the target attribute but also for increased entropy in protected attributes. Our experiments show that GroupCART achieves fairer models without data transformation and with minimal performance degradation. Furthermore, the method supports customizable weighting, offering a smooth and flexible trade-off between predictive performance and fairness based on user requirements. These results demonstrate that algorithmic bias in decision tree models can be mitigated through multi-task, fairness-aware learning. All code and datasets used in this study are available at: https://github.com/anonymous12138/groupCART.
format Preprint
id arxiv_https___arxiv_org_abs_2504_12587
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Software Engineering Principles for Fairer Systems: Experiments with GroupCART
Peng, Kewen
Zhuo, Hao
Yang, Yicheng
Menzies, Tim
Machine Learning
Software Engineering
Discrimination-aware classification aims to make accurate predictions while satisfying fairness constraints. Traditional decision tree learners typically optimize for information gain in the target attribute alone, which can result in models that unfairly discriminate against protected social groups (e.g., gender, ethnicity). Motivated by these shortcomings, we propose GroupCART, a tree-based ensemble optimizer that avoids bias during model construction by optimizing not only for decreased entropy in the target attribute but also for increased entropy in protected attributes. Our experiments show that GroupCART achieves fairer models without data transformation and with minimal performance degradation. Furthermore, the method supports customizable weighting, offering a smooth and flexible trade-off between predictive performance and fairness based on user requirements. These results demonstrate that algorithmic bias in decision tree models can be mitigated through multi-task, fairness-aware learning. All code and datasets used in this study are available at: https://github.com/anonymous12138/groupCART.
title Software Engineering Principles for Fairer Systems: Experiments with GroupCART
topic Machine Learning
Software Engineering
url https://arxiv.org/abs/2504.12587