Quantifying Group Fairness in Community Detection

2025-04-15

AI Summary

This work addresses group-level unfairness in community detection arising from structural inequities (e.g., race, gender). We formally define multiple quantifiable group fairness metrics and establish an attribute-aware fairness evaluation framework. Methodologically, we systematically assess 12 state-of-the-art community detection algorithms using synthetic benchmarks (LFR, ABCD, HICH-BA) and real-world networks, integrating graph structural analysis with statistical fairness testing. Our empirical analysis reveals a pervasive trade-off between detection performance and fairness across diverse settings. Notably, Infomap and Significance emerge as balanced algorithms—exhibiting both high fairness and strong community detection capability. This study delivers the first reproducible, multi-scenario fairness evaluation standard for community detection, accompanied by practical guidelines for fair algorithm selection and deployment.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Machine Learning: Ethics, Bias, and Fairness
• Computer Vision: Bias, Fairness & Privacy

Application Category

• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems

Abstract

Understanding community structures is crucial for analyzing networks, as nodes join communities that collectively shape large-scale networks. In real-world settings, the formation of communities is often impacted by several social factors, such as ethnicity, gender, wealth, or other attributes. These factors may introduce structural inequalities; for instance, real-world networks can have a few majority groups and many minority groups. Community detection algorithms, which identify communities based on network topology, may generate unfair outcomes if they fail to account for existing structural inequalities, particularly affecting underrepresented groups. In this work, we propose a set of novel group fairness metrics to assess the fairness of community detection methods. Additionally, we conduct a comparative evaluation of the most common community detection methods, analyzing the trade-off between performance and fairness. Experiments are performed on synthetic networks generated using LFR, ABCD, and HICH-BA benchmark models, as well as on real-world networks. Our results demonstrate that the fairness-performance trade-off varies widely across methods, with no single class of approaches consistently excelling in both aspects. We observe that Infomap and Significance methods are high-performing and fair with respect to different types of communities across most networks. The proposed metrics and findings provide valuable insights for designing fair and effective community detection algorithms.

Problem

• Assessing fairness in community detection algorithms
• Evaluating trade-offs between performance and fairness
• Proposing metrics for fair community detection methods

Innovation

• Novel group fairness metrics for community detection
• Comparative evaluation of fairness-performance trade-offs
• Infomap and Significance methods excel in fairness