A Descriptive and Predictive Analysis Tool for Criminal Data: A Case Study from Brazil

Yan Andrade, Matheus Pimenta, Gabriel Amarante, Antônio Hot Faria, Marcelo Vilas-Boas, João Paulo da Silva, Felipe Rocha, Jamicel da Silva, Wagner Meira Jr., George Teodoro, Leonardo Rocha, Renato Ferreira

Abstract: Addressing the challenge of crime is crucial for governing bodies, requiring informed strategies. This article examines the underutilization of detailed criminal data, collaborating with the Military Police of Minas Gerais, Brazil. We propose a new methodology, materialized in a tool, that is able to transform raw data into strategic information for public security decision-making. The tool evaluation unfolds in three phases: characterizing the data, a descriptive analysis of a real case study, and a predictive analysis. This work highlights the untapped potential in detailed criminal data, emphasizing the pivotal role of precise analysis in deciphering complex dynamics. Collaborating with law enforcement aims to bridge the gap between data abundance and actionable insights for effective public security strategies.