Prioritizing of offenders in networks

This paper reports work that builds upon several years of experimentation using forensic psychology guided exploratory techniques from artificial intelligence, statistics and spatial statistics. Our central aim is the development of decision support systems for crime prevention and detection, and this paper presents a novel algorithm that incorporates geographical information, frequency and recency of criminal activity directly into the ‘betweenness ’ metric of social network analysis. The algorithm is ad hoc, and design decisions are presented, alongside the operational use by police forces of such an algorithm, namely as a means for prioritizing of offenders in large networks. The data presented is from the crime of burglary from dwelling houses.