JurisConsulto: retrieval in jurisprudencial text bases using juridical terminology

  • Authors:
  • Tânia C. D'Agostini Bueno;Christiane Gresse von Wangenheim;Eduardo da Silva Mattos;Hugo Cesar Hoeschl;Ricardo M. Barcia

  • Affiliations:
  • Graduate Program - Production Engineering, Federal University of Santa Catarina, Brazil;Graduate Program - Production Engineering, Federal University of Santa Catarina, Brazil;Graduate Program - Production Engineering, Federal University of Santa Catarina, Brazil;Graduate Program - Production Engineering, Federal University of Santa Catarina, Brazil;Graduate Program - Production Engineering, Federal University of Santa Catarina, Brazil

  • Venue:
  • ICAIL '99 Proceedings of the 7th international conference on Artificial intelligence and law
  • Year:
  • 1999

Quantified Score

Hi-index 0.00

Visualization

Abstract

In the legal domain, jurisprudence has an important role as a juridical source; its decisions support the application of the Law to a concrete case. The problem is that Brazilian Courts produce an enormous amount of decisions every year, turning these text sources larger every time and forcing juridical professionals to spend more time in the search for a relevant decision. Sophisticated AI techniques are needed to minimize searches time and improves the quality and appropriateness of the retrieved information. This paper describes a case-based approach for the intelligent retrieval of jurisprudencial texts. The approach enables the retrieval of adequate texts with characteristics similar to information supplied by the user in natural language. New documents are automatically included into the knowledge base by extracting relevant information. In order to enable the processing of informal textual knowledge in natural language, a controlled vocabulary and a juridical thesaurus based on common juridical terminology, is integrated into the retrieval and extraction process. The approach is based on sentences to criminal proceedings in the domain of the Brazilian Right.