Characterizing User Behavior: The Interplay Between Mobility Patterns and Mobile Traffic

TL;DR

This work addresses understanding how individual mobility patterns interact with mobile traffic by leveraging week-long XDR data at 30-minute granularity across Chile. It introduces a dynamic, feature-rich user characterization framework spanning 13 traffic and mobility features, and develops a lightweight Markov inference approach that captures bidirectional dependencies while preserving population heterogeneity. The methodology demonstrates that mobility is more predictable from traffic than vice versa, identifies key mobility features that drive traffic profiles, and validates robust generalization across urban contexts with about 94–97% mobility inference accuracy in use cases. The results enable realistic integrated datasets and offer a privacy-conscious, interpretable alternative to deep-learning methods for cross-domain user behavior modeling with potential applications in adaptive services and synthetic data generation.

Abstract

Mobile devices have become essential for capturing human activity, and eXtended Data Records (XDRs) offer rich opportunities for detailed user behavior modeling, which is useful for designing personalized digital services. Previous studies have primarily focused on aggregated mobile traffic and mobility analyses, often neglecting individual-level insights. This paper introduces a novel approach that explores the dependency between traffic and mobility behaviors at the user level. By analyzing 13 individual features that encompass traffic patterns and various mobility aspects, we enhance the understanding of how these behaviors interact. Our advanced user modeling framework integrates traffic and mobility behaviors over time, allowing for fine-grained dependencies while maintaining population heterogeneity through user-specific signatures. Furthermore, we develop a Markov model that infers traffic behavior from mobility and vice versa, prioritizing significant dependencies while addressing privacy concerns. Using a week-long XDR dataset from 1,337,719 users across several provinces in Chile, we validate our approach, demonstrating its robustness and applicability in accurately inferring user behavior and matching mobility and traffic profiles across diverse urban contexts.

Figures (12)

• Figure 1: Mobile users' traffic behavior distributions
• Figure 2: Users' mobility behavior features distributions
• Figure 3: (a)-(b) Traffic features by mobility profile; (c)-(e) Mobility features distribution by traffic profile.
• Figure 4: Dependency between in-step traffic and mobility features of users in Santiago
• Figure 5: Accuracy of inferring mobility from traffic behavior