An open-source implementation of geographic profiling methods for serial crime analysis

The rgeoprofile R package was developed to implement functions for the analysis of serial crime incidents. Geographic profiling is an investigative technique that utilizes the spatial relationship of a connected series of crime incidents to determine or predict the most probable area of offender residence or anchor point. If successfully used as a decision support system, criminal geography can be used to help law enforcement agencies strategically target certain areas for inquiry or prioritize suspects through a narrowed search window. As an open-source platform, the rgeoprofile package contains several rapid reproducible models of spatial analysis using either centrographic or distance decay functions to predict the offender's home base. An open-source approach results in transparent analyses with no-cost availability for agencies. Additionally, since both mathematical models and investigator heuristics have been shown to provide viable options for criminal geographic profiling, a software package, which integrates different solutions to the geographic profiling problem was needed. Finally, the article demonstrates the various geographic profiling methods in a case study of the Boston Strangler to illustrate the advantages of each approach.