I worked for some time on a research project on international criminology. Mostly what I learned was that crime data tend to be pretty thorough in developed countries and really quite bad in the developing world. Luckily a group of researchers (Frank et al. 2011) got a hold of five years of real-world crime data from British Columbia, Canada and used it for some great visualizations. Their paper presents a model of crime locations and attractors to those hotspots based on distance to the offender’s home.
What’s going on here? The visualization of this data demonstrates their conclusion that, “crime is not randomly distributed, [but] rather it exhibits clear spatial patterns.” (take a look at Brantingham and Brantingham 1981). Looking at where offenders live, they chart their movement to crime attractors using those nice vector lines.
In their model, crime generators draw people who are not otherwise criminals, but see an opportunity too good to pass up, while attractors lure career offenders to known opportunities. Despite the seemingly ‘haphazard’ distribution of crimes in the area they studied (a couple of medium-sized urban centers), they speculate that, “an underlying pattern should be present.”
You know what I think about pattern recognition. I like this already.
They hypothesize that crimes would be committed along routes between the offenders homes and activity node locations. Incorporating directionality into their original model, they visualize their findings using vectors:
Beyond looking pretty neat, I like how they delve into the qualitative aspects of their model. “We are interested,” they write, “in what attracted the offender to offend at that specific location based on the distance between the Crime Location and Attractor.”
All in all a good paper and a great example of a simple yet effective visualization. Keep an eye out on the Journal of Artificial Societies and Social Simulation (website with full issues here) for more interesting stuff.
P.L. Brantingham and P.J. Brantingham, Notes on the geometry of crime, In P.J. Brantingham and P.L. Brantingham (Eds.),Environmental Criminology, Sage Publications, Beverly Hills, CA, (1981).
Richard Frank, Vahid Dabbaghian, Andrew Reid, Suraj Singh, Jonathan Cinnamon and Patricia Brantingham. (2011). Power of Criminal Attractors: Modeling the Pull of Activity Nodes. Journal of Artificial Societies and Social Simulation 14(1) 6:http://jasss.soc.surrey.ac.uk/14/1/6.html.
- Predicting crime before it happens (flowingdata.com)