Project Archive

Criminal Activity Vectors - The Effect of Directionality on the Activity Space of Offenders

To determine if there is directionality to crime with respect to home location. Do offenders move in a specific or random direction on their journey to crime? This research shows that yes, offenders have a directionality preference.

Criminal Activity Vectors - The Attractiveness of City Centers

Evaluate the attractiveness of the city center for offenders.

Criminal Activity Vectors

According to previous research, offenders have a directionality preference when moving about in their environment. Using clustering techniques, this theory is tested using real life crime data from across British Columbia, Canada. Once directionality was established, a unique clustering technique, based on K-Means clustering and modified for angles, was applied to find the number of activity paths for each offender.

Criminal Activity Vectors - Predicting the Activity Space Nodal Points of Offenders

According to Crime Pattern Theory, individuals all have routine daily activities located in several nodes that are used for different purposes, such as home, work or shopping. As people move between these nodes, their familiarity with the spatial area around the nodes, as well as between nodes, increases. Offenders have the same spatial movement patterns and awareness spaces as regular people, hence it is likely that an offender will commit the crimes in their own awareness space and on the way to one of their nodes. This idea is used to predict the location of the nodes within the awareness space of offenders. The activities of 57,962 offenders who were charged or charges were recommended against them were used to test this idea by mapping their offense locations with respect to their home locations to determine directionality. Once directionality to crime was established, a unique clustering technique, based on K-Means clustering and path-prediction, was used to find the most likely location of one of the nodes of the offender. It was found that, by selecting the right clustering parameters, offenders tend to move towards central shopping areas, and commit crimes along the way.

Criminal Activity Vectors - Predicting the Full Activity Space of Offenders

This research is the culmination of the above research into predicting the full activity space of offenders. This is done by analyzing the directionality movement of offenders, and predicting the nodal points, then reconstructing the full activity space of offenders.

Intelligent Agents

Masterminds 1 & 2:

ICURS works with and supports the development of the Mastermind project that explores offender's mobility model, formation of the activity space, target selection, development of social networks and the creation of hot spots. This is done using Abstract State Machine (ASM) formalism and abstraction principles as a common core for linking the different aspects and views.

Mastermind2 has been modified to allow the development of simulation models for the testing of theories. The researcher enters formal rules to describe the theory and basic contextual (backcloth) conditions. Mastermind2 then simulates the results of the actions/ activities of the rules.

Police Management Information System (PMIS)

ICURS is working on incorporating Homicide, Victim and UCR2 datasets into the other systems housed at ICURS. This dataset, which covers the entire province of BC, is still under development.

Power of Criminal Attractors - Modeling the Pull of Activity Nodes

We use offenders' home locations and the locations of their crimes to define directional and distance parameters. Using these parameters we apply mathematical structures to define rules by which different models may behave to investigate the influence of activity nodes on the spatial distribution of crimes in crime neutral areas. The findings suggest an increasing likelihood of crime as a function of geometric angle and distance from an offender's home location to the site of the criminal event. Implications of the results are discussed.

Process Modelling of Criminal Justice System

The process model that simulates the interrelationships between crime trends, policing, and court systems.


A developing algorithm for linking the geographic and cognitive psychology sides of criminology research with a prototype topology algorithm that joins local urban areas together using rules that define similarity between adjacent small units of analysis. This research looks for a pattern modeling approach that uses topology to spatially identify the concentrations of crime and their crisp breaks and gradual blending into adjacent areas using the basic components: interior, boundary and exterior.

Voronoi Diagrams

Traditional cluster analysis is usually based on k-means and uses a Euclidean space. Crime data is reported on streets by address. The streets form a network within a Euclidean space. Improved algorithms are being developed to make it possible to use Dijkstra distances between lots in urban areas to develop a spatial tessellation that reflects true travel distances. This algorithm is being used to select nodal points like shopping malls and to determine their likely catchment areas for sales and for crimes.