Analysis of keyword extraction methods for legal document classification

Authors

  • Matheus S. Marinato Postgraduate Program in Computer and Systems Engineering (PECS), State University of Maranhão (Uema)
  • Ewaldo E. C. Santana Postgraduate Program in Computer and Systems Engineering (PECS), State University of Maranhão (Uema)
  • Antonio Fernando Lavareda Jacob Junior Postgraduate Program in Computer and Systems Engineering (PECS), State University of Maranhão (Uema)

DOI:

https://doi.org/10.5335/rbca.v16i2.15528

Keywords:

Keyword Extraction, Procedural Classification, Judicial Efficiency

Abstract

Given the significant increase in judicial demand and the growing scarcity of resources in the Brazilian legal system, it becomes imperative to invest in new technological applications. A critical area for analysis is the processing of legal cases, which still involves several manually conducted stages, including the initial screening. In this context, this research aims to analyze the techniques of keyword extraction and verify which extraction technique can help in the classification of legal documents. To this end, two databases were used, one containing 86 cases and the other with 3,543 jurisprudence, both belonging to the Judiciary Court of Maranhão. The results showed that the Bertopic and YAKE extraction methods obtained better performance for the legal data. It is concluded that these models have the potential to be used to provide greater speed in the procedural distribution.

Downloads

Download data is not yet available.

Published

2024-08-05

Issue

Section

Original Paper

How to Cite

[1]
2024. Analysis of keyword extraction methods for legal document classification. Brazilian Journal of Applied Computing. 16, 2 (Aug. 2024), 88–96. DOI:https://doi.org/10.5335/rbca.v16i2.15528.