Revista Colombiana de Computación (Jun 2017)
Using Hotspots Algebra to analyze crime events and their urban context
Abstract
Hotspots analysis is essential in the criminology field and quite important in decisions making for police agencies because it permits the enhancement of allocation of police resources for timely and adequate actions. There exist different techniques for analysis and generation of hotspots, limited by its difficulty to consider other urban and demographic factors that could be the cause of the emergence of these hotspots or their influence over other factors. On the other hand, it is also difficult and unusual to analyze hotspots in a temporal context. This paper seeks, through spatial-temporal operations based in hotspots, to go beyond of classical crime analysis with hotspots, by looking for the spatial influence of other spatial factors over and analyzing also their relationship in a temporal context. The paper initially presents and analyze the performance of various techniques for hotspot generation and determines that STAC technique from CrimeStat is the more suitable for the proposed objective. Then, it defines a hotspot algebra allowing the combined study of crime and spatial factors affecting it and/or been affected by crime events. Temporal analysis includes the hotspots generation for days of the week and/or months in a year. In this way it is possible to study causality relationships and/or correlations among the studied phenomena and the spatial-temporal environment where crime occurs. Therefore, it is possible to define and apply informed actions, primarily concerning the allocation of police resources. Finally, it shows some application examples, thematic as well as temporal analysis, of hotspot algebra using crime data from Bogota for the years 2011 to 2013; finally, some future works in the subject are proposed.
