Knowledge Discovery Lab

Predicting Criminal Activity


Criminal activities are driven by underlying determinants and unevenly distributed over space. The concept of hotspots is widely used to analysis the spatial characteristics of crimes. But existing methods usually identify hotspots based on an arbitrary user-defined threshold with respect to the occurrences of a target crime without considering underlying controlling factors. In this study we introduce a new data mining model { Hotspots Optimization Tool (HOT) { to optimize the identification of crime hotspots. The key component of HOT, Geospatial Discriminative Patterns (GDPatterns), captures the difference between two classes in the spatial dataset. Super Patterns, which are clustered GDPatterns based on similarity measures, are used to demonstrated the distribution and structure of crime determinants. Using a dataset collected from a northeastern city in the United States, we demonstrate that the HOT model is a useful tool in optimizing the discovery of crime hotspots, and super patterns are capable of describing controlling factors of crime, which will help domain scientists further their understanding of the underlying reasons of criminal activities.


Understanding the Spatial Distribution of Crime Based on Its Related Variables Using Geospatial Discriminative Pattern

D. Wang*, W. Ding, H. Lo, M. Morabito, P. Chen, J. Salazar, and T. Stepinski, Computers, Environment and Urban Systems, Elsevier, 2013

Crime Hotspot Mapping Using the Crime Related Factors--A Spatial Data Mining Approach

D. Wang*, W. Ding, H. Lo, T. Stepinski, J. Salazar, and M. Morabito, Applied Intelligence, Springer, 2012

Optimization of Criminal HotSpots Based on Underlying Crime Controlling Factors Using Geospatial Discriminative Pattern

D. Wang*, W. Ding, T. Stepinski, J. Salazar*, and M. Morabito, The 25th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, June, Dalian, China, 2012

Crime Forecasting Using Data Mining Techniques

C. Yu*, M. Ward*, M. Morabito, and W. Ding, The 4th Workshop on Data Mining Case Studies and Practice Prize, Vancouver, Canada, December, 2011

Empirical Discriminative Tensor Analysis for Crime Forecasting

Y. Mu*, W. Ding, M. Morabito, D. Tao, the 5th International Conference on Knolwege Science, Engineering and Management, Irvine, CA, December, 2011

Individuals Involved

Dawei Wang, Yang Mu, Wei Ding, Henry Lo, Chung-Hsien Yu

Comments (0) Trackbacks (0)

No comments yet.

Leave a comment

You must be logged in to post a comment.

Trackbacks are disabled.