Network-Based Space-Time Scan Statistics for Detecting Micro-Scale Hotspots

Events recorded in urban areas are often confined by the micro-scale geography of street networks, yet existing spatial–analytical methods do not usually account for shortest-path distance networks. We propose space–time NetScan, a new spatial–temporal analytical method with improved accuracy detecting patterns concentrations across space and time. It extends notion scan-statistic-type search window measuring space-time along networks order to detect events at street-address level high accuracy. Performance tests synthetic data demonstrate that NetScan outperforms location, shape, size duration hotspots. An empirical study drug-related incidents shows how can improve our understanding crime. Aside from some abrupt one-off incidents, many hotspots form recurrent hotbeds, implying crimes tend persist specific problem places.