Urban Anomaly Analytics: Description, Detection, and Prediction

Urban anomalies may result in loss of life or property if not handled properly. Automatically alerting their early stage even predicting before happening is great value for populations. Recently, data-driven urban anomaly analysis frameworks have been forming, which utilize big data and machine learning algorithms to detect predict automatically. In this survey, we make a comprehensive review the state-of-the-art research on analytics. We first give an overview four main types anomalies, traffic anomaly, unexpected crowds, environment individual anomaly. Next, summarize various datasets obtained from diverse devices, i.e., trajectory, trip records, CDRs, sensors, event data, social media surveillance cameras. Subsequently, survey issues detecting techniques presented. Finally, challenges open problems as discussed.