FairGB: A Fair Granular-Ball Generation Method for Data Classification

With the widespread application of data-driven classifiers in high-risk domains, group fairness has become an important research focus. However, most existing methods rely on model constraints or data reweighting, which may limit interpretability or distort the original data distribution. Granular-ball computing (GBC), as a structured and interpretable learning framework, provides a natural foundation for incorporating group fairness into data partitioning. Based on this insight, we propose a $\textbf{Fair}$ $\textbf{G}$ranular-$\textbf{B}$all $\textbf{G}$eneration framework (FairGBG), which employs fair clustering to maintain balanced proportions of sensitive groups within each granular ball (GB), thereby enhancing within-ball group fairness. Theoretical analysis shows that FairGBG can preserve high GB purity while satisfying group fairness requirements. Furthermore, we introduce a $\textbf{Fair}$ $\textbf{G}$ranular-$\textbf{B}$all-based data $\textbf{F}$air $\textbf{C}$lassification method (FairGBFC), which leverages fair GBs to improve classification fairness. Experiments on multiple benchmark datasets demonstrate that, compared with existing GB generation methods, FairGBG can generate locally fair GBs. Moreover, compared with state-of-the-art fairness-aware baselines, FairGBFC achieves a superior trade-off between accuracy and fairness, effectively mitigating bias while preserving high utility. Our code is publicly available at https://github.com/jamesandai/FairGB.

Machine learning models are increasingly used to support important decisions in areas such as lending, education, healthcare, and law. However, these models may treat different groups of people unfairly, for example by producing less favorable outcomes for certain gender or racial groups. This paper proposes a new method to make such decision systems fairer while keeping them accurate and easier to understand. The main idea is to organize the data into many small local regions, where each region contains similar samples and maintains a balanced representation of different sensitive groups. These local regions are then used to guide the final prediction. Because each region has clear information about its label consistency and group balance, the method can make decisions in a more transparent way instead of relying only on a black-box model. Experiments on several real-world datasets show that the proposed method can reduce unfair differences between groups while maintaining strong prediction performance. Overall, this work provides a more interpretable and practical approach for building fairer machine learning classifiers in high-stakes decision-making scenarios.