Spatial-temporal patterns of aggressive behaviors. A case study Bogotá, Colombia

Understanding of crime patterns is paramount important for citizen's security planning. In particular, the comprehension of the Spatio-temporal dynamics related to aggressive behaviors is fundamental for deploying policial resources and devising mitigation actions. Currently, a significant number of approaches to find patterns, characterize dynamics, and predict crime have been proposed in state of the art. However, the operation of these approaches is strongly adapted to the specific conditions of a city. In this paper, we propose a novel approach to finding spatio-temporal crime patterns in the city of Bogotá (Colombia), particularly aggressive behaviors reported to the emergency line. We characterize aggressive behaviors through rhythm and tempo based on the theory of routine activity. Finally, we show that the dynamics of aggressive behaviors in the city are shared by several spatial units in which specific strategies can be applied to mitigate it.

I. Introduction

Recently, there has been an increase in the development of computational models to predict and characterize crime [1]–[5]. Among the motivations is to take prevention actions, optimize available resources, improve indicators and the perception of security, and define more effective public policies based on safety [4]. However, differences in environmental conditions, culture, size of cities, and type of crime change behavior patterns and crime dynamics, which means that many