Spatio-temporal Anomaly Detection in Intelligent Transportation Systems

• DOI: 10.1016/j.procs.2019.04.117
• Publication Date: 2019-05-21T20:48:31Z
• Authors: Mai H. Hassan; Ali Tizghadam; Alberto Leon-Garcia
• Keywords: Anomaly (physics); Temporal database

With the increased availability of real-time data streams in different domains comes opportunity using this to provide valuable insights into performance systems generating such data. In paper, we are proposing an anomaly detection method be applied on road traffic intelligent transportation systems. The proposed scheme is based multi-channel singular spectrum analysis (MSSA), and aims characterize spatio-temporal properties network. By simultaneously analyzing spatial temporal attributes network, able detect contextual collective anomalies that otherwise undetectable only or techniques. This indicated through results shown experiments section.