Understanding Urban Behavior: Information Theory Insights from WhatsApp Traffic Analysis

Instant messaging represents the most popular digital service worldwide, and WhatsApp is the most used app of this kind. By studying WhatsApp traffic, we can gain major insight into both human behavior and network infrastructure needs. This work presents a unique analysis of mobile networks using WhatsApp uplink traffic. We apply information theory metrics such as Shannon Permutation Entropy, Statistical Complexity, and the Causality Complexity-Entropy Plane to understand specific network patterns, usage behaviors, and areas where users are more likely to engage in online conversations. We also demonstrate how these metrics can be used as features for machine learning techniques such as the K-means algorithm, showing they can be used to identify regions with similar patterns in the WhatsApp network traffic.