摘要

This study investigates patterns of micro-scale concentrations of different types of crime using the network distance in the spatial, temporal and spatial-temporal dimensions to enable an accurate description of the micro-scale geospatial variation of crime incidents. It applies a recently developed hotspot detection method that uses a network-based space-time search window technique. The method is refined by adopting the false discovery rate controlling procedure for the multiple testing problem. Empirical analysis uses individual street-address records of robbery, burglary, drug and vehicle theft incidents in a high-crime neighbourhood of Chicago in the year 2000. The study revealed a fine-scale, street-address-level space-time signature for each type of crime. Drugs and robbery formed stable space-time hotspots in specific locations, highlighting their recurrent nature. Burglary was characterised by a small set of short-term outbursts across space and time, and vehicle thefts showed little sign of concentrations. Comparing these results against their spatial signature helped identify different types of hotspots such as persistent warm spots and a hotspot consisting of a short-term outburst.The result demonstrates the significance of the street-level analysis from the microscopic perspective, which can help form a more focused policing tactic.

  • 出版日期2015-5-4