DETERMINANTES E PREDIÇÃO DE CRIMES DE HOMICÍDIOS NO BRASIL: UMA ABORDAGEM DE APRENDIZADO DE MÁQUINA

Abstract

Throughout history, organized societies have attempted to prevent crime by various approaches, and the social justification for understanding what features associated with crime is their use to achieve effective public policies against such illegal activities. In this context, the objective of this work is to identify the economic, social and demographic determinants of homicide crimes in Brazil. And, as a secondary objective, the prediction of the level of crimes in national territory. As a methodology, this is a quantitative study where Regression Tree, Random Forest, Boosting and K-Nearest Neighbors (K-NN) methods were used as alternative tools to traditional linear models, such as regression via ordinary least squares. The data analyzed indicated that among the 31 covariates used, 9 had the greatest impacts on violence at the national level, such as the size of the young population, basic sanitation, total population size, economically active population, urban population, GDP , female heads in the family, poor people between 0 and 14 years old and the proportion of people earning up to half a minimum wage, where each factor was discussed according to the economic literature of the crime. In addition, the Random Forest model explained, on average, 82% of the variability of homicide crimes at the national level. We believe that this approach helps to produce more robust responses on the effects of social, economic, and demographic factors on crime and is therefore a new tool for guiding policymakers.