Data Analysis in Predicting Crime : Predictive Policing

摘要

Crime is a constant in our society. Even though there is a huge amount of data available about crime it is hard to find the best solution to prevent it. However, through analysis of historical crime data it is possible to predict the risk of crime in time and space. This article presents the results of a literature review that gives insights into what predictive policing is, its theoretical foundations and its application in a real-world context. It also gives an overview of known predictive policing softwares – Crime Antecipation System, PreCobs, PredPol and Hunchlab along with their main characteristics. These systems allow a more efficient allocation of resources of law enforcement agencies and informs crime prevention strategies. Nevertheless, these solutions are new and come with limitations, which makes the evaluation of its impact in crime rates difficult.