Comparação de Técnicas de Clusterização em Documentos de Texto em Português
DOI:
https://doi.org/10.5753/isys.2025.5029Keywords:
Clusterização de texto, BERT, BERTimbau, K-Means, Ligação simples, Modelo de Mistura GaussianaAbstract
Lidar com a enorme quantidade de dados de texto no mundo digital é um desafio complexo. Uma abordagem eficaz para enfrentá-lo é através da técnica de agrupamento de documentos de texto. Este estudo avaliou o desempenho de três algoritmos de clustering — K-Means, Single Linkage e Gaussian Mixture Model (GMM) — na categorização de artigos de notícias em português brasileiro utilizando o modelo BERTimBau, variação em português do modelo BERT, para pré-processamento. As métricas utilizadas foram acurácia, F1-score, índice Rand e coeficiente de Jaccard. Os resultados dessas métricas indicaram que o Single Linkage obteve a melhor performance global, superando K-Means e GMM na maioria dos critérios avaliados.
Downloads
Referências
Aggarwal, C. C. and Reddy, C. K. (2013). Data Clustering: Algorithms and Applications. Data Mining and Knowledge Discovery Series. Chapman & Hall/CRC, Boca Raton, FL.
AI, G. (2018). Open sourcing bert: State-of-the-art pre-training for natural State-of-the-art pre-training for natural language processing. Disponível em: [link]. Acesso em: 27 mar. 2025.
Ayub, M., Ghazanfar, M., Khan, T., et al. (2020). An effective model for Jaccard coefficient to increase the performance of collaborative filtering. Arab Journal of Science and Engineering, 45:9997–10017.
Campello, R. (2007). A fuzzy extension of the Rand index and other related indexes for clustering and classification assessment. Pattern Recognition Letters, 28(7):833–841.
Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. (2019). BERT: Pre-training of deep bidirectional transformers for language understanding. In Burstein, J., Doran, C., and Solorio, T., editors, Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 4171–4186, Minneapolis, Minnesota. Association for Computational Linguistics.
Fayyad, U., Piatetsky-Shapiro, G., and Smyth, P. (1996). From data mining to knowledge discovery in databases. AI Magazine, 17(3):37.
Ferreira, W. (2023). Desvendando padrõees de dados: Uma jornada pelos métodos de clusterização (k-means, gaussian mixture, ...). Disponível em: [link]. Acesso em: 27 mar. 2025.
Francisco Jáñez-Martino, Rocío Aláiz-Rodríguez, V. G.-C. E. F. E. A. (2023). Classifying spam emails using agglomerative hierarchical clustering and a topic-based approach. FREEDOM, 139:110226.
Khishigsuren Davagdorj, Ling Wang, M. L. V.-H. P. K. H. R. N. T.-U. (2022). Discovering thematically coherent biomedical documents using contextualized bidirectional encoder representations from transformers-based clustering. 19(10):5893.
Lima, A. (2021). Introdução aos transformers. Disponível em: [link]. Acesso em: 27 mar. 2025.
Manpreet Kaur, . V. (2021). An improved k-means based text document clustering using artificial bee colony with support vector machine. 8(7).
Mohammad Alhawarat, M. O. H. (2018). Revisiting k-means and topic modeling, a comparison study to cluster Arabic documents. 6:42740.
Souto Moreira, L., Machado Lunardi, G., de Oliveira Ribeiro, M., Silva, W., and Paulo Basso, F. (2023). A study of algorithm-based detection of fake news in Brazilian election: Is BERT the best? IEEE Latin America Transactions, 21(8):897–903.
Souza, F., Nogueira, R., and Lotufo, R. (2020a). BERTimbau: Pretrained BERT models for Brazilian Portuguese. In Cerri, R. and Prati, R. C., editors, Intelligent Systems, pages 403–417, Cham. Springer International Publishing.
Souza, F., Nogueira, R., and Lotufo, R. (2020b). BERTimbau: Pretrained BERT models for Brazilian Portuguese. In Cerri, R. and Prati, R. C., editors, Intelligent Systems, pages 403–417, Cham. Springer International Publishing.
Subakti, A., Murfi, H., and Hariadi, N. (2022). The performance of BERT as data representation of text clustering. Journal of Big Data, 9(1):15.
Tan, P.-N., Steinbach, M., and Kumar, V. (2014). Introduction To Data Mining. Pearson, New York.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., and Polosukhin, I. (2017). Attention is all you need. In Advances in Neural Information Processing Systems, pages 6000–6010, Cambridge, MA. MIT Press.
Velankar, A., Patil, H., and Joshi, R. (2023). Mono vs multilingual BERT for hate speech detection and text classification: A case study in Marathi. In El Gayar, N., Trentin, E., Ravanelli, M., and Abbas, H., editors, Artificial Neural Networks in Pattern Recognition, pages 121–128, Cham. Springer International Publishing.
Weiss, S. M., Indurkhya, N., and Zhang, T. (2015). Fundamentals of Predictive Text Mining. Springer International Publishing, London, second edition edition.
Wu, S. and Dredze, M. (2020). Are all languages created equal in multilingual BERT? In Gella, S., Welbl, J., Rei, M., Petroni, F., Lewis, P., Strubell, E., Seo, M., and Hajishirzi, H., editors, Proceedings of the 5th Workshop on Representation Learning for NLP, pages 120–130, Online. Association for Computational Linguistics.
Downloads
Published
Como Citar
Issue
Section
Licença
Copyright (c) 2025 iSys - Revista Brasileira de Sistemas de Informação
Este trabalho está licenciado sob uma licença Creative Commons Attribution 4.0 International License.
