Anomaly Detection in Partially Observed Traffic Networks

TL;DR

This paper introduces a Bayesian hierarchical model for detecting anomalies in traffic networks with incomplete observations, enabling effective identification of unusual activity through statistical goodness-of-fit tests.

Contribution

It presents a novel hierarchical Bayesian approach for anomaly detection in partially observed traffic networks, improving detection accuracy even with model misspecification.

Findings

Model performs well on simulated data

Outperforms existing methods on real datasets

Effective even with model misspecification

Abstract

This paper addresses the problem of detecting anomalous activity in traffic networks where the network is not directly observed. Given knowledge of what the node-to-node traffic in a network should be, any activity that differs significantly from this baseline would be considered anomalous. We propose a Bayesian hierarchical model for estimating the traffic rates and detecting anomalous changes in the network. The probabilistic nature of the model allows us to perform statistical goodness-of-fit tests to detect significant deviations from a baseline network. We show that due to the more defined structure of the hierarchical Bayesian model, such tests perform well even when the empirical models estimated by the EM algorithm are misspecified. We apply our model to both simulated and real datasets to demonstrate its superior performance over existing alternatives.