Optimization of criminal hotspots based on underlying crime controlling factors using geospatial discriminative pattern

Authors:
Dawei Wang;Wei Ding;Tomasz Stepinski;Josue Salazar;Henry Lo;Melissa Morabito
Affiliations:
Department of Computer Science, University of Massachusetts, Boston;Department of Computer Science, University of Massachusetts, Boston;Department of Geography, University of Cincinnati;Department of Computer Science, Rice University;Department of Computer Science, University of Massachusetts, Boston;College of Liberal Arts, University of Massachusetts, Boston
Venue:
IEA/AIE'12 Proceedings of the 25th international conference on Industrial Engineering and Other Applications of Applied Intelligent Systems: advanced research in applied artificial intelligence
Year:
2012

Citing 3
Cited 0

Efficient mining of emerging patterns: discovering trends and differences

KDD '99 Proceedings of the fifth ACM SIGKDD international conference on Knowledge discovery and data mining
Discovering Frequent Closed Itemsets for Association Rules

ICDT '99 Proceedings of the 7th International Conference on Database Theory
Mining statistically important equivalence classes and delta-discriminative emerging patterns

Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining

Quantified Score

Hi-index	0.00

Visualization

Abstract

Criminal activities are unevenly distributed over space. The concept of hotspots is widely used to analyze the spatial characters of crimes. But existing methods usually identify hotspots based on an arbitrary user-defined threshold with respect to the number of a target crime without considering underlying controlling factors. In this study we introduce a new data mining model --- Hotspots Optimization Tool (HOT) --- to identify and optimize crime hotspots. The key component of HOT, Geospatial Discriminative Patterns (GDPatterns), which capture the difference between two classes in spatial dataset, is used in crime hotspot analysis. Using a real world dataset of a northeastern city in the United States, we demonstrate that the HOT model is a useful tool in optimizing crime hotspots,and it is also capable of visualizing criminal controlling factors which will help domain scientists further understanding the underlying reasons of criminal activities.