Assessing Predictive Models for Fairness Based on Movement Patterns

2026-05-22

AI Summary

Existing approaches to spatial fairness evaluation are largely confined to static, single-location representations—such as residential addresses—and overlook how individuals’ mobility patterns across regions may influence model fairness. This work introduces a novel framework that, for the first time, incorporates individual movement trajectories into spatial fairness analysis. The proposed paradigm integrates multi-resolution geographic partitioning, trajectory mapping, and spatial scan statistics to effectively detect prediction unfairness induced by mobility behaviors. Evaluated on thousands of synthetic datasets, the method demonstrates high accuracy and robustness, precisely identifying affected individuals while achieving stable fairness-localization trade-offs across multiple spatial scales.

Abstract

Assessing the spatial fairness of predictive models involves establishing whether they are statistically penalizing (favoring) individuals associated with certain geographical locations. Literature on this topic makes the fundamental assumption that each individual is assigned to a single geographical location (e.g., place of residence). However, fairness with respect to the set of locations where one has been, i.e., their movement patterns over different regions, also matters when fairness is considered. Consequently, we argue that it is necessary to generalize the notion of spatial fairness to also include movement patterns, leading to the novel problem of assessing predictive models for fairness relative to the movements of individuals. To deal with this problem, we propose an approach that first associates the movements of individuals to certain geographic regions, considering multiple spatial partitions with different resolutions and alignments, and then employs a suitable spatial scan statistic to assess whether a predictive model is fair based on movement patterns. In the experimental evaluation, we study the performance of our approach over thousands of synthetic unfair datasets, showing that it is effective at detecting this new type of unfairness and at retrieving the set of objects treated unfairly, while localization performance exhibits a consistent multi-resolution trade-off.

Problem

• spatial fairness
• movement patterns
• predictive models
• geographical locations
• fairness assessment

Innovation

• movement patterns
• spatial fairness
• predictive models
• spatial scan statistic
• multi-resolution analysis