On Predicting Sociodemographics from Mobility Signals

2025-11-06

AI Summary

Predicting demographic attributes from mobile trajectory data faces challenges including weak attribute–mobility correlations and poor generalization across domains. To address these, we propose directed mobility graphs to encode high-order, structured movement patterns of individuals. We further design a multi-task learning framework that jointly predicts age, gender, income, and household composition, thereby enhancing cross-scenario and cross-temporal distributional robustness. Additionally, we introduce a confidence–accuracy-balanced visualization diagnostic tool to improve model interpretability and enable principled uncertainty quantification. Extensive experiments demonstrate that our approach significantly outperforms state-of-the-art baselines under data scarcity and distribution shift, achieving superior prediction accuracy and robustness. This work establishes a new paradigm for trustworthy utilization of passive mobility data in urban transportation planning.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Inferring sociodemographic attributes from mobility data could help transportation planners better leverage passively collected datasets, but this task remains difficult due to weak and inconsistent relationships between mobility patterns and sociodemographic traits, as well as limited generalization across contexts. We address these challenges from three angles. First, to improve predictive accuracy while retaining interpretability, we introduce a behaviorally grounded set of higher-order mobility descriptors based on directed mobility graphs. These features capture structured patterns in trip sequences, travel modes, and social co-travel, and significantly improve prediction of age, gender, income, and household structure over baselines features. Second, we introduce metrics and visual diagnostic tools that encourage evenness between model confidence and accuracy, enabling planners to quantify uncertainty. Third, to improve generalization and sample efficiency, we develop a multitask learning framework that jointly predicts multiple sociodemographic attributes from a shared representation. This approach outperforms single-task models, particularly when training data are limited or when applying models across different time periods (i.e., when the test set distribution differs from the training set).

Problem

• Predicting sociodemographics from mobility data with weak relationships
• Improving model interpretability while maintaining predictive accuracy
• Enhancing generalization across different contexts and time periods

Innovation

• Higher-order mobility descriptors from directed graphs
• Visual diagnostic tools for model confidence and accuracy
• Multitask learning framework for shared representation across attributes