Project combines data science and sociology to map crime

With the support of FAPESP, the research is developing mathematical and computational tools that will address issues related to public safety in an innovative way.

Agência FAPESP* – Over the next few years, a group of 22 Brazilian researchers and two foreign collaborators intend to develop mathematical and computational tools that will innovatively address issues related to crime, impunity and the legitimacy of the institutions in charge of public security in the country.

Selected by the FAPESP eScience and Data Science Research Program, the project “Criminality, insecurity and legitimacy: a transdisciplinary approach” brings together specialists in data science, computer science and social sciences.

The team brings together the experience and representativeness of members of two Research, Innovation and Dissemination Centers (RIDCs) supported by FAPESP: the Center for the Study of Violence (NEV) and the Center for Mathematical Sciences Applied to Industry (CeMEAI).

“This project is a new proposal for joint work between the RIDCs that began in 2016. We can highlight at least three research fronts: legitimacy, impunity, and urban and criminal patterns. And in addition to these three fronts, the Thematic Project also has important lines of action that include a data portal to organize and analyze information and databases related to crime and the multidisciplinary training of students and researchers,” explains the researcher in charge, Luis Gustavo Nonato.

Objectives and results

The methodologies and analytical tools that will be developed in this project will allow complex studies to be conducted in the areas of crime and security in order to generate and prove new hypotheses.

“We hope that the results will serve as a basis for the design of evidence-based public policies, enabling public security agents to better plan actions to reduce and prevent specific crimes. The studies should also identify the main factors associated with fear and insecurity among the population, so that targeted actions can be taken to increase the sense of security in urban agglomerations, thus contributing to improving quality of life and sustainability,” explains Nonato.

The researcher draws attention to the ethical issue that has been one of the main concerns since the beginning of the project: “Almost nothing has been done so far to try to understand how sensitive the data used by the models that make predictions are, and, above all, to what extent the results pointed out can in some way discriminate against certain social genders, social classes and even urban regions, specific localities. All of this can be highly impacted by the models. How far can we go and how can we go? How far are we going to go and what path are we going to take to avoid, as much as possible, impacting unfavorably on populations with serious social problems? We have to be very careful about this.”

Another goal of the project is to bring the university closer to companies and government bodies that work in or deal with issues related to public safety.

NEV’s scientific coordinator, Sérgio Adorno, points to the current era of transdisciplinarity: “Data science has a lot to teach us and we have a lot to teach them. And the most important thing is that in the end we can make a diagnosis of the system as a whole and contribute to a reform of the criminal justice system in the medium and long term. I think we have to be very clear about our role as scientists and intellectuals in producing knowledge, pointing out the problems and potentially suggesting ways forward and solutions.”

* With information from CeMEAI, a FAPESP Research, Innovation and Dissemination Center.