Unsupervised Ensemble Clustering for Fake News Detection in Brazilian Portuguese using Contextual and Traditional Embeddings

• Nathalia Raquel Bianchini UNESP
• Jean Alex Firmino Pinhata UNESP
• Alex Marino Gonçalves de Almeida FATEC Ourinhos
• Rodrigo Capobianco Guido UNESP

DOI: https://doi.org/10.5753/eniac.2026.26701

Resumo

The automated detection of fake news in Brazilian Portuguese often relies on supervised learning, which may be constrained by the scarcity of labeled data. In this study, we evaluate an unsupervised ensemble clustering framework using the Fake.br Corpus and compare traditional baselines (TF-IDF, Doc2Vec, FastText) with contextual embeddings (BERTimbau). Our results suggest that BERTimbau tends to group deceptive narratives into clusters with higher internal consistency and to separate legitimate news into two thematic groupings. These results indicate that dense contextual vectors could have potential for unsupervised disinformation detection in this language.

Referências

Bellman, R. (1961). Adaptive Control Processes: A Guided Tour. Princeton University Press, Princeton, NJ.

Bojanowski, P., Grave, E., Joulin, A., and Mikolov, T. (2017). Enriching Word Vectors with Subword Information. Transactions of the Association for Computational Linguistics, 5:135–146. DOI: 10.1162/tacl_a_00051.

Colomina, C., Margalef, H. S., Youngs, R., and Jones, K. (2021). The impact of disinformation on democratic processes and human rights in the world. Study, European Parliament. DOI: 10.2861/59161.

Das, K., Kamlish, and Abid, F. (2024). Advancements in Word Embeddings: A Comprehensive Survey and Analysis. Proceedings of the Pakistan Academy of Sciences: A. Physical and Computational Sciences, 61(3):227–245. DOI: 10.53560/PPASA(61-3)842.

Fayyad, U., Piatetsky-Shapiro, G., and Smyth, P. (1996). From Data Mining to Knowledge Discovery in Databases. AI Magazine, 17(3):37–54. DOI: 10.1609/aimag.v17i3.1230.

Han, J., Kamber, M., and Pei, J. (2011). Data Mining: Concepts and Techniques. Morgan Kaufmann, Waltham, MA, 3rd edition. DOI: 10.1016/C2009-0-61819-5.

Jurafsky, D. and Martin, J. H. (2026). Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition with Language Models. 3rd edition. Online manuscript released January 6, 2026.

Le, Q. and Mikolov, T. (2014). Distributed Representations of Sentences and Documents. In Proceedings of the 31st International Conference on Machine Learning, pages 1188–1196. PMLR. Available at: [link]. Accessed: April 21, 2026.

McInnes, L., Healy, J., and Melville, J. (2020). UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction. DOI: 10.48550/arXiv.1802.03426.

Monteiro, R. A., Santos, R. L. S., Pardo, T. A. S., de Almeida, T. A., Ruiz, E. E. S., and Vale, O. A. (2018). Contributions to the Study of Fake News in Portuguese: New Corpus and Automatic Detection Results. In Computational Processing of the Portuguese Language, pages 324–334, Cham. Springer International Publishing. DOI: 10.1007/978-3-319-99722-3_33.

Newman, N., Ross Arguedas, A., Robertson, C. T., Nielsen, R. K., and Fletcher, R. (2025). Digital News Report 2025. Reuters Institute for the Study of Journalism. Available at: [link]. Accessed: April 17, 2026.

Paixão, M., Lima, R., and Espinasse, B. (2020). Fake News Classification and Topic Modeling in Brazilian Portuguese. In 2020 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT), pages 427–432. DOI: 10.1109/WIIAT50758.2020.00063.

Reimers, N. and Gurevych, I. (2019). Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 3982–3992, Hong Kong, China. Association for Computational Linguistics. DOI: 10.18653/v1/D19-1410.

Rousseeuw, P. J. (1987). Silhouettes: A Graphical Aid to the Interpretation and Validation of Cluster Analysis. Journal of Computational and Applied Mathematics, 20:53–65. DOI: 10.1016/0377-0427(87)90125-7.

Shah, A., Shah, H., Bafna, V., Khandor, C., and Nair, S. (2025). Validation and Extraction of Reliable Information Through Automated Scraping and Natural Language Inference. Engineering Applications of Artificial Intelligence, 147:110284. DOI: 10.1016/j.engappai.2025.110284.

Souza, F., Nogueira, R., and Lotufo, R. (2020). BERTimbau: Pretrained BERT Models for Brazilian Portuguese. In Intelligent Systems, pages 403–417, Cham. Springer International Publishing. DOI: 10.1007/978-3-030-61377-8_28.

Sparemberger, R. and da Silva, A. C. E. S. (2021). O IMPACTO DAS FAKE NEWS NO PROCESSO ELEITORAL BRASILEIRO. Revista Reflexão e Crítica do Direito, 9(2):251–277. Available at: [link]. Accessed: March 29, 2026.

Trevilato, L. A. (2025). Análise Comparativa de Modelos de Representação de Texto e Métodos de Aprendizado de Máquina na Classificação de Notícias Falsas em Português. Undergraduate thesis, Universidade Federal de Uberlândia. Available at: [link]. Accessed: March 29, 2026.

Vieira, C., Souza, J., and Cavalcanti, G. (2025). Detecção de Fake News em Português: Análise Comparativa entre Métodos de Representação em Português, Inglês e Multilíngues. In Anais do XIV Brazilian Workshop on Social Network Analysis and Mining, pages 187–199, Porto Alegre, RS, Brasil. SBC. DOI: 10.5753/brasnam.2025.9062.