Discovering Spatial-Temporal Indication of Crime Association (STICA)

• In Event: New Approaches to Understanding Property Crime

Abstract

This paper proposes an approach of Spatial-Temporal Indication of Crime Association (STICA) to explore contributing factors to crime through discerning spatial zones with distinct temporal crime patterns. Combining Kernel Density Estimation and K-Median-Centers Clustering, a specific implementation of the STICA approach is applied to burglary in the DP peninsula, Southern China. The empirical results show that both stable and seasonal factors affecting crime can be identified with the graphical indication of disparities in temporal crime patterns of different zones, while corresponding spatial coverages of these factors in STICA results can enlighten the understanding of their interactions in generating crime patterns, especially about how temporally transient and spatially global factors can produce a locally crime-ridden zone through the mediation of stable factors. Also, STICA results can reveal the spatially contextual effects of stable factors, which are of great value to improve understanding and modelling crime patterns. As demonstrated, STICA approach is effective in exploring contributing factors to crime and has shown great potential for providing a new vision in place-based crime research.

Authors

• Chao Jiang, Sun Yat-sen University

• Lin Liu, University of Cincinnati, Sun Yat-sen University

• Xiaoxing Qin, Sun Yat-sen University

• Suhong Zhou, Sun Yat-sen University