Revista Opinião Jurídica (May 2020)

COMPLIANCE TO LEGAL PRECEDENTS AND REPETITIVE LEGAL CASES MANAGEMENT WITH MACHINE LEARNING

  • Isabela Cristina Sabo,
  • Aires José Rover

DOI
https://doi.org/10.12662/2447-6641oj.v18i28.p69-93.2020
Journal volume & issue
Vol. 18, no. 28
pp. 69 – 93

Abstract

Read online

Brazilian Judiciary is presented as one of the focus of discussion on the implementation of technologies related to Artificial Intelligence (AI), from which comes the machine learning (ML). In this context, the research problem is presented: is it possible that machine learning contributes to the functioning of the legal precedents system? Starting from an initial positive hypothesis, and using the inductive method, the article presents a study of two cases, with data collection about two procedural institutes: one about the non-compliance to legal precedents, the “Reclamação” [sic]; and another about the techniques for management of repetitive cases, the “Incidente de Resolução de Demandas Repetitivas (IRDR)” [sic]. In order to contextualize the theme and objectives of the study, the research presents the evolution of the concept of jurisdiction, referring to the changes brought by the Brazilian Code of Civil Procedure (2015), and still the notions of E-Judiciary and e-jurisdiction combined with the functionalities of AI, as well as related work using ML. Thus, the proposed ML applications are based on the identification of these two procedural institutes, related to the operation of the legal precedence system, which can be improved with the implementation of the technology to achieve a model of intelligent civil jurisdiction.

Keywords