Automatic Classification of Combined Classifiers of Initial Petitions

• Matheus Serrão Marinato UEMA
• Antonio F. L. Jacob Junior UEMA
• Fábio M. F. Lobato UEMA / UFOPA
• Omar A. C. Cortes UEMA / IFMA

DOI: https://doi.org/10.5753/bresci.2022.223296

Abstract

The increase in judicial demand and the scarcity of resources is expressive, harming the service and agility of the Brazilian legal system. Thus, it was perceived the need to invest in new technological applications to ensure the smooth running of the processes. Thus, the correct classification of petitions facilitates the categorization of the process in its initial phase, avoiding future repetitive demands that deal with the same situation. In this context, this research aims to use combined machine learning to classify initial requests, streamlining the initial process. To this end, a database was built with 1,787 models of petitions from various portals with legal content. The results show that the combined classifier presented the best performance, being 90% of accuracy, 91% of precision, 87% of recall and 89% of F1-Score. It is concluded that the model has the potential to be used to provide greater speed in the procedural distribution.

Keywords: Initial petition, machine learning, Classification

References

Almeida Neto, M., Moura, V., Bandeira, J., Freitas, P., and Fagundes, R. (2018). Um modelo de inferência para a classificação de resultados processuais da justiça estadual. Macabéa-Revista Eletrônica do Netlli, 3(3).

Breiman, L. (2001). Random forests. Machine learning, 45(1):5–32.

Campestrini, H. and Florence, R. C. (2000). Como redigir petição inicial. Saraiva Educação SA, São Paulo, SP.

Chen, T. and Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, pages 785–794.

Cirqueira, D., Almeida, F., Cakir, G., Jacob, A., Lobato, F., Bezbradica, M., and Helfert, M. (2020). Explainable sentiment analysis application for social media crisis management in retail. In 4th International Conference on Computer-Human Interaction Research and Applications Volume 1: WUDESHI-DR.

Cirqueira, D., Fontes Pinheiro, M., Jacob, A., Lobato, F., and Santana, Á. (2018). A literature review in preprocessing for sentiment analysis for brazilian portuguese social media. In 2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI).

CNJ (2021). Conselho Nacional de Justiça. Justiça em números 2021. CNJ, Brasília.

Faraco, F. M. (2020). Modelo de conhecimento baseado em tópicos de acórdãos para suporte à análise de petições iniciais. Master’s thesis, Universidade Federal de Santa Catarina, Florianópolis, SC, Brasil.

Hollanda, Y. R. d. and Leite, F. G. d. F. (2020). Petição Inicial: uma análise à luz de teorias bakhtinianas. Macabéa-Revista Eletrônica do Netlli, 9(4):292–308.

Justiça Digital (2021). Tjam automatiza classificação de petições intermediárias no portal e-saj, https://justicadigital.com/tjam-ia-peticionamento/, visitado em 28/11/2021.

Leme, B. (2021). Classificação automática de documentos de características econômicas para defesa jurídica. Master’s thesis, Universidade de São Paulo, São Paulo, SP, Brasil.

Liang, M., Chang, T., An, B., Duan, X., Du, L., Wang, X., Miao, J., Xu, L., Gao, X., Zhang, L., et al. (2021). A stacking ensemble learning framework for genomic prediction. Frontiers in genetics, 12.

Palaia, N. (2000). Técnica da petição inicial. Saraiva Educação SA, São Paulo, SP.

Sousa, G. N. d., Almeida, G. R., and Lobato, F. (2019). Social network advertising classification based on content categories. In International Conference on Business Information Systems, pages 396–404. Springer.

Sousa, R. N. d. (2019). Minerjus: solução de apoio à classificação processual com uso de inteligência artificial. Master’s thesis, Universidade Federal do Tocantins, Palmas, TO, Brasil.

Yi, H.-C., You, Z., Wang, M.-N., Guo, Z.-H., Wang, Y., and Zhou, J.-R. (2020). Rpi-se: a stacking ensemble learning framework for ncrna-protein interactions prediction using sequence information. BMC Bioinformatics, 21.

Zhou, Z. H. (2009). Ensemble Learning, Encyclopedia of Biometrics, volume 1. Springer, Berlin, Germany.