Integrating near repeat and social network approaches to analyze crime patterns

Crime does not occur randomly or uniformly across time, space, or social groups. The near-repeat phenomenon states that the risk of repeat victimization would increase at the nearby locations within a certain time period. The empirical evidence of near-repeat patterns has been widely reported in many different crime types across the world. Meanwhile, the growing ability to connect anywhere and anytime has led to the increasing use of social network analytics. In particular, social network analytics have been applied to the studies of criminal networks. This paper integrates near repeat and social network approaches on exploring spatiotemporal crime patterns. The spatiotemporal burglary data in five boroughs (the Bronx, Brooklyn, Staten Island, Manhattan and Queen) of NYC were analyzed and compared based on average clustering coefficient, degree centrality, closeness centrality, and closeness centrality. Furthermore, the implications and limitations of the findings are presented.