Improving Geographic Profiling through Commuter/Marauder Prediction

Abstract

One of the most talked about new technologies in the field of crime analysis is geographic profiling, an investigative methodology that uses the locations of a connected series of crimes to determine the most probable offender anchor point. Although geographic profiling software has been around since the early 1990s the science of geographic profiling has languished considerably. Much of the research on geographic profiling over the last 15 years has focused on developing new software platforms and providing case study analysis, rather than on critically evaluating issues of effectiveness and accuracy. This lack of research has created a void in which many questions about geographic profiling still exist, such as: is there an effective method for determining if a crime series can be profiled?, and what role does opportunity structure play in profile results? It is because of this void that the current research has been conducted. The focus of this research is on developing basic spatial analysis methods that can improve the overall accuracy and effectiveness of geographic profiles. In particular, this research seeks to develop simple spatial diagnostics that can be used to assist in deciding which cases are capable of being profiled as well as improving overall profile accuracy. In addition to research results this presentation will include a discussion of the implications of these findings for practitioners of geographic profiling.