Crime and Its Social Context: Analysis Using the Self-Organizing Map

Abstract

Data mining and visualization techniques show their value in various domains but have not been broadly applied to the study of crime, which is in demand of an instrument to efficiently and effectively analyze available data. The purpose of this study is to apply the Self-Orgamizing Map (SOM) to mapping countries with different situations of socio-economic development. Supplemented by other methods, including Scatter Counter for attribute selection, and nearest neighbor search, discriminant analysis and decision trees for obtaining comparable results, the SOM is found to be a useful tool for mapping criminal phenomena through processing of multivariate data.