Spatio-temporal Anomaly Detection in Traffic Data

Spatio-temporal data mining has received much attention in recent years in many industrial and financial applications. Anomaly detection has also become an important problem. The detection of anomalies in spatio-temporal traffic data is an important problem in the data mining and knowledge discovery community. In this paper, we first investigate multiple types of traffic data and extract different features from each type of the data. Then, we combine grid partition on the basis of Local Outlier Factor (LOF) algorithm and develop a grid-based LOF algorithm to detect the abnormal area in Beijing. Finally, we conduct extensive experiments on real-world trip data including taxi and bus data. And experimental demonstrate the effectiveness of our proposed approach.