Discovering significant situational profiles of crime occurrence by modeling complex spatial interactions

Abstract The joint influence of different facilities plays an important role in understanding situational profiles of crime incidents. While spatial conjunctive analysis of case configurations (CACC) is widely used to explore situational profiles of crime, complex spatial interactions (e.g., spatial autocorrelation of crime incidents, spatial interactions among multiple facilities) have not been fully considered. Drawing on previous environmental criminology research, this study extends the CACC by modeling complex spatial interactions between crime and facilities. First, the spatial interaction range between single facility and crime incidents is quantitatively measured by the cross-type pair-correlation function. Then, the relationship between different situational profiles is modeled, and a “bottom-up” strategy is applied to generate potential situational profiles. Finally, significant situational profiles are selected via Monte Carlo testing by modeling the complex structure of crime incidents. The effectiveness of the proposed approach is evaluated by both a synthetic and real dataset. The experimental result manifests that the proposed approach can effectively eliminate the influence of independent facilities and more accurately identify those significant situation profiles. The discovered significant situational profiles have a positive guiding effect in understanding the spatial context for crime occurrence, thereby facilitating crime prevention.