IBAD: Interpretable Behavioral Anomaly Detection on Human Mobility Data

2026-06-14

AI Summary

This work addresses the limited interpretability of existing location-based anomaly detection methods by proposing IBAD, a novel framework that detects anomalies at the level of behavioral semantics rather than raw geographic coordinates. IBAD leverages Latent Dirichlet Allocation (LDA) to extract a small set of globally interpretable behavioral templates and constructs a hierarchical self-supervised model that represents individual behaviors as mixtures over these templates. The approach introduces behavior prototypes that generalize across diverse geographic and demographic contexts and establishes the first behavioral stitching benchmark for evaluating controlled anomalies. Experimental results demonstrate that everyday human behaviors can be effectively decomposed into semantically meaningful and interpretable templates, and that IBAD achieves strong generalization, high interpretability, and robust anomaly detection performance on both real-world and synthetic datasets.

Abstract

Human mobility appears highly diverse, yet much of a person's daily mobility can be explained by a small set of recurring behavioral templates, such as commuting, school-centered activities, caregiving, nightlife, or errand patterns. We present \texttt{IBAD} (\underline{I}nterpretable \underline{B}ehavioral \underline{A}nomaly \underline{D}etection), a framework that learns interpretable daily mobility templates and represents each individual as a distribution over mixtures of these templates. Rather than focusing on specific locations, IBAD characterizes activities that individuals perform across locations. This approach first discovers global behavioral templates using Latent Dirichlet Allocation (LDA), then employs a hierarchical self-supervised model to learn normal behavior of individuals from their soft behavioral templates. We also introduce a \emph{splicing benchmark} that creates controlled behavioral mismatches between an individual's historical profile and injected mobility patterns. Experiments on real-world and synthetic datasets show that daily behavior can be effectively decomposed into a small number of interpretable templates. Crucially, we show that the learned behavioral archetypes \emph{transfer} across distinct geographic and demographic contexts. Furthermore, IBAD maintains a robust competitive performance across all settings. For reproducibility purposes, the code is accessible at ~\href{https://github.com/USC-InfoLab/IBAD}{https://github.com/USC-InfoLab/IBAD}.

Problem

• Behavioral Anomaly Detection
• Human Mobility
• Interpretable Models
• Mobility Templates
• Anomaly Detection

Innovation

• Interpretable Anomaly Detection
• Behavioral Templates
• Latent Dirichlet Allocation
• Self-supervised Learning
• Human Mobility