According to Crime Pattern Theory, individuals all have routine daily activities which require frequent travel between several nodes, with each used for various purposes, such as their home, work, or shopping location. As people move about, their familiarity with the spatial areas around, and in between, the nodes increases, eventually forming their Activity Space. Offenders have similar spatial movement patterns and Activity Spaces as non-offenders, hence, according to theory, an offender will commit the crimes in their own Activity Space. The goal of this Dissertation is to introduce a novel approach, called Directionality-based Activity Space Creator (DASC), to reconstruct the Activity Space of offenders. Before this reconstruction however, first the crime locations for each individual were analyzed to establish directionality preference on the individual level, after which the directionality preferences were analyzed at the city level to establish aggregate patterns. Once the existence of directionality preference had been established, these preferences were analyzed across various crime types and cities to show robustness. These preferences were then used to detect the individual Paths in the Activity Spaces. Information about all Paths were then used to detect the Nodes in the region, which were then assigned to individuals, completing their Activity Space. To test the accuracy of this model, the Activity Spaces of 322 repeat offenders within the City of Surrey, Canada, were reconstructed and used to recommend likely suspects for new crimes. Recommendation was based on the crime’s proximity to the Activity Spaces of the offenders. The higher the actual offender was within the recommended set, the more accurate the model. Results indicate that by reconstructing the Activity Spaces, recommendations could be made which were five to seven times as accurate as the naïve selection.
Copyright is held by the author.
The author granted permission for the file to be printed and for the text to be copied and pasted.
Supervisor or Senior Supervisor
Thesis advisor: Andresen, Martin
Member of collection