Spatiotemporal Modeling and Monitoring for Network Application
DescriptionIn recent decades, the increasingly interest in scientific analysis at a system level and the ever-growing capabilities for high-throughput data collection in various fields has fueled the explosion of network data. However, few research has been conducted on spatiotemporal analysis of social network. Due to the temporal dependency of social network data and complex cross-dimensional dependence among entities, it is a challenging task to detect anomalies effectively in a dynamic network system. In this research, we will develop an effective, reliable, and credible monitoring framework tailored to multidimensional data streams over networks. We focus on developing a scan-based approach for detecting anomalies in dynamic networks. In particular, we v.-ill (i) develop modeling and monitoring methods for spatiotemporal surveillance in dynamic social network environment; (iii) evaluate and validate the proposed methods using case studies, and compare their performance with the state of art methods.
|Effective start/end date||1/09/18 → …|