TopoFair: Linking Topological Bias to Fairness in Link Prediction Benchmarks

TL;DR

This paper introduces a benchmarking framework that links topological biases in social graphs to fairness in link prediction, revealing how structural biases affect fairness interventions and their generalization.

Contribution

It proposes a new framework for evaluating fairness in link prediction by considering structural biases, including a taxonomy, a graph generation method, and empirical analysis.

Findings

Fairness interventions are sensitive to structural biases beyond homophily.

The framework enables controlled variation of biases to test model robustness.

Empirical results highlight the importance of structural considerations in fairness evaluations.

Abstract

Graph link prediction (LP) plays a critical role in socially impactful applications, such as job recommendation and friendship formation. Ensuring fairness in this task is thus essential. While many fairness-aware methods manipulate graph structures to mitigate prediction disparities, the topological biases inherent to social graph structures remain poorly understood and are often reduced to homophily alone. This undermines the generalization potential of fairness interventions and limits their applicability across diverse network topologies. In this work, we propose a novel benchmarking framework for fair LP, centered on the structural biases of the underlying graphs. We begin by reviewing and formalizing a broad taxonomy of topological bias measures relevant to fairness in graphs. In parallel, we introduce a flexible graph generation method that simultaneously ensures fidelity to real-world graph patterns and enables controlled variation across a wide spectrum of structural biases. We apply this framework to evaluate both classical and fairness-aware LP models across multiple use cases. Our results provide a fine-grained empirical analysis of the interactions between predictive fairness and structural biases. This new perspective reveals the sensitivity of fairness interventions to beyond-homophily biases and underscores the need for structurally grounded fairness evaluations in graph learning.