Network Model for Crime Forecasting

Abstract

The theories about crime and correction have their inception in the eighteenth century, highly influenced by the anthropological thoughts emerging during the age of Enlightenment. Throughout the decades, the criminological studies observed their sociological essence encompassing practices from other scientific fields to explain the more contemporary questions, becoming Criminology therefore an inherently interdisciplinary science. The adoption of concepts from Exact Sciences is a recent moving, originating it a novel research area, called Computational Criminology, which employs procedures from Applied Mathematics, Statistics and Computer Science to provide original or enhanced solutions to such questions. One of the most prominent tasks brought by this rising field is crime prediction, which attempts to uncover potential targets for future police intervention and also help solving offenses already committed. The present Research Internship Abroad project thus analyzes the usage of Bayesian networks for predictive policing, continuing the research initiated in the course of the Regular Program in Brazil, when a comparative analysis about the available algorithms to learn a Bayesian network structure purely from data was established. From the results previously obtained, the studies at Monash University, Australia, aim to answer the main challenge of the doctoral research linked to this proposal, that is, to investigate the viability of the application of Petri nets to build a novel score-based structure learning algorithm, by working together with leading international Bayesian network specialists. A collaboration with the Australian Federal Police in the development of a new artificial intelligence procedure is also expected, fulfilling it the social contribution intended by the original project in Brazil.