Performance Evaluation of BERT Models for Automated Police Report Classification in Marabá-PA

Authors

  • Marcílio Douglas Silva Marques Federal University of Southern and Southeastern Para
  • Reginaldo C. dos Santos Filho
  • Hugo Pereira Kuribayashi
  • Adam D. Ferreira dos Santos
  • Anderson da Silva Soares

DOI:

https://doi.org/10.5335/rbca.v18i1.17029

Keywords:

Applications of AI, Classification, Machine Learning, Police Reports, Violence

Abstract

This work addresses the development of a classifier for police reports from the city of Marabá-PA, employing data mining techniques and fine-tuning of pre-trained Large Language Models (LLMs), such as Bidirectional Encoder Representations from Transformers (BERT) and its Portuguese-adapted version, BERTimbau. The evaluation of the models indicates that the BERT base and BERTimbau transformers achieved overall accuracies of approximately 90% and 92%, respectively, in experiments conducted with test data. These results demonstrate the feasibility of using LLMs for the automatic classification of police reports, offering potential to enhance criminal data analysis and contribute to more efficient, data-driven public security policies.

Downloads

Download data is not yet available.

Published

2026-04-29

Issue

Section

Original Paper

How to Cite

[1]
2026. Performance Evaluation of BERT Models for Automated Police Report Classification in Marabá-PA. Brazilian Journal of Applied Computing. 18, 1 (Apr. 2026), 68–80. DOI:https://doi.org/10.5335/rbca.v18i1.17029.