An Auto Encoder Network Based Method for Abnormal behavior detection

With the continuous development of wireless communication technology and positioning technology, the acquisition of spatiotemporal trajectory data has become easier. Through data mining, discovering valuable knowledge hidden in trajectory data can be widely used in activities recommendation, urban planning, military and other fields, which has important research value. So this paper studies the behavior mining technology for trajectory data, the main contents is abnormal behavior detection based on deep learning. A trajectory anomaly detection algorithm based on variational self-encoding network is proposed. This method uses GRU as the basic unit of encoding and decoding, the reconstruction probability as the anomaly score. The proportion of suspected anomalies is introduced to adaptively adjust the abnormality judgment threshold. From the experiment, the accuracy and recall rate of the anomaly detection algorithm are both higher than 90%, and the real-time detection efficiency is high, which can meet the needs in actual scenarios.