Science for Development Center: State Prosecutors, Innovation and Technology - São Paulo (PROINTE-SP)

• Grant number: 24/01027-4
• Support Opportunities: Research Grants - Science Centers for Development
• Start date: July 01, 2024
• End date: June 30, 2029
• Field of knowledge: Applied Social Sciences - Law - Public Law
• Principal Investigator: Paulo Furquim de Azevedo
• Grantee: Paulo Furquim de Azevedo
• Host Institution: Instituto de Ensino e Pesquisa (Insper). São Paulo , SP, Brazil
• Pesquisadores principais: Hedibert Freitas Lopes ; Rogério Bastos Arantes
• Associated researchers: Diana Loureiro Paiva de Castro ; Fabio Jose Ayres ; Guilherme da Franca Couto Fernandes de Almeida ; Henrique Yu Jiunn Wang ; Igor Moraes Rocha ; Julio Adolfo Zucon Trecenti ; Tiago Fernandes Tavares

Abstract

The State Attorney's Office of São Paulo (PGE-SP) deals with a series of repetitive litigation themes, estimated at two million legal cases, in addition to monitoring judgments in the State Court of Auditors of São Paulo (TCE-SP). However, there are no systematized data on the outcomes of these judgments or the costs they generate for the institution. In response to this, in partnership with the Institute of Teaching and Research - Insper -, the proposal of this research center is to produce technologies for generating information capable of assisting PGE-SP in managing legal cases, making decisions for cost assessment, and the convenience of changes in administrative procedures. Thus, by simplifying certain procedures due to the existence of already established jurisprudence and thereby avoiding the costs of excessive and unnecessary judicialization, the idea is to produce positive effects in Public Administration and the Judiciary, reducing the impact on public accounts financed by the whole society. Following the latest artificial intelligence techniques, the research intends to combine knowledge of law with mathematics, statistics, computer science, and linguistics for the development of these technologies and procedural improvements capable of allowing PGE-SP to achieve the previously exposed objectives. For this, the object of the research will be the judicial decisions involving strategic themes in which PGE-SP operates, which will be analyzed with various Machine Learning and Data Science techniques to assess the dynamics involving procedural processing and costs. In the first phase of the research, the general data of the actions will be organized, with the objective of grouping them according to similar characteristics. The second phase will aim to understand the characteristics of each of these groups in terms of procedural movements, allowing the measurement of essential statistical data such as duration time, probability of success, among others. Finally, techniques of Natural Language Processing (NLP) and Artificial Intelligence (AI) will be applied with the intention of establishing textual similarities between the decisions, automating document generation, and studying the legal issues, laws, and jurisprudence involved. The products of the research will be made available in public repositories, becoming a perennial contribution to the field of Social-Science in general, and Law in particular. (AU)