Emojis and Hate Speech in the Brazilian Context: An Analysis Across Different Social Media Platforms

• Thúlio M. O. Fernandes UFV
• Paula T. M. Gibrim UFV
• Julio C. S. Reis UFV

DOI: https://doi.org/10.5753/brasnam.2026.23590

Abstract

The widespread adoption of social media in Brazil has established platforms such as Instagram, YouTube, and Twitter/X as central spaces for interaction, but also as channels for the dissemination of hate speech. While computational approaches for its detection have advanced, the role of emojis remains underexplored in the national context. This work investigates the relationship between emoji usage and hate speech in Brazilian Portuguese (PT-BR) using a dataset of over 30,000 labeled messages from different platforms, analyzed with Natural Language Processing (NLP) techniques. The results indicate that some emojis are consistently associated with hate speech, while others exhibit context-dependent variability. These findings highlight the potential of emojis as complementary signals in automated detection models. Warning! This work and the referenced data contain examples of potentially offensive and hateful language.

References

Aldunate, N. and González-Ibáñez, R. (2017). An integrated review of emoticons in computer-mediated communication. Frontiers in psychology, 7:2061.

Althobaiti, M. J. (2022). Bert-based approach to arabic hate speech and offensive language detection in twitter: exploiting emojis and sentiment analysis. International Journal of Advanced Computer Science and Applications, 13(5).

Bertot, J. C., Jaeger, P. T., and Hansen, D. (2012). The impact of polices on government social media usage: Issues, challenges, and recommendations. Government information quarterly, 29(1):30–40.

Biere, S., Bhulai, S., and Analytics, M. B. (2018). Hate speech detection using natural language processing techniques. Master Business AnalyticsDepartment of Mathematics Faculty of Science.

Braga, M. L. P., Nakamura, F. G., and Nakamura, E. F. (2020). Criação e caracterização de um corpus de discurso sexista em português. In Brazilian Workshop on Social Network Analysis and Mining (BraSNAM), pages 97–107.

Caetano, J., Guimarães, S., Araújo, M. M., Silva, M., Reis, J. C., Silva, A. P., Benevenuto, F., and Almeida, J. M. (2022). Characterizing early electoral advertisements on twitter: A brazilian case study. In International Conference on Social Informatics, pages 257–272.

de Freitas Melo, P., Kansaon, D., Couto, J. M., Reis, J. C., and Benevenuto, F. (2025). A sticker is worth a thousand words: Characterizing the use and abuse of stickers on whatsapp political groups in brazil. In Proc. of the Int’l AAAI Conference on Web and Social Media, volume 19, pages 1210–1223.

de Oliveira, F. R., Reis, V. D., and Ebecken, N. F. F. (2024). Detecting hate speech on brazilian social media: New dataset and analysis. In Ibero-Latin American Congress on Computational Methods in Engineering (CILAMCE).

De Pelle, R. P. and Moreira, V. P. (2017). Offensive comments in the brazilian web: a dataset and baseline results. In Brazilian Workshop on Social Network Analysis and Mining (BRASNAM), pages 510–519.

de Santana, V. F., Melo-Solarte, D. S., de Almeida Neris, V. P., de Miranda, L. C., and Baranauskas, M. C. C. (2009). Redes sociais online: desafios e possibilidades para o contexto brasileiro. In Seminário Integrado de Software e Hardware (SEMISH), pages 339–353.

Fortuna, P., da Silva, J. R., Wanner, L., Nunes, S., et al. (2019). A hierarchically-labeled portuguese hate speech dataset. In Proceedings of the third workshop on abusive language online, pages 94–104.

Fortuna, P. and Nunes, S. (2018). A survey on automatic detection of hate speech in text. ACM Computing Surveys, 51(4):85:1–85:30.

Grosz, P. G., Greenberg, G., De Leon, C., and Kaiser, E. (2023). A semantics of face emoji in discourse. Linguistics and Philosophy, 46(4):905–957.

Grover, V. and Banati, H. (2024). An attention approach to emoji focused sarcasm detection. Heliyon, 10(17):e36398.

Guimarães, S., Silva, M., Caetano, J., Araújo, M., dos Reis, J. C. S., da Silva, A. P. C., Benevenuto, F., and Almeida, J. M. (2022). Análise de propagandas eleitorais antecipadas no twitter. In Brazilian Workshop on Social Network Analysis and Mining (BraSNAM).

Ibrohim, M. O., Setiadi, M. A., and Budi, I. (2019). Identification of hate speech and abusive language on indonesian twitter using the word2vec, part of speech and emoji features. In Proceedings of the International Conference on Advanced Information Science and System (AISS), pages 1–5.

Leite, J. A., Silva, D., Bontcheva, K., and Scarton, C. (2020). Toxic language detection in social media for brazilian portuguese: New dataset and multilingual analysis. In Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing, pages 914–924.

Moreira, L. S., Gibrim, P. T. M., Rocha, L., and Reis, J. C. S. (2026). Anatomy of data repositories for the analysis and detection of toxicity in portuguese. In Proceedings of the International Conference on Computational Processing of Portuguese (PROPOR) - Vol. 1. Association for Computational Lingustics.

Pinto, S. L., Campolina, J. J., Sena, J. P. M., Félix, G., Ferreira, L. N., and Reis, J. C. (2024). Caracterização e predição de usuários tóxicos no twitter/x durante as eleições brasileiras de 2022. In Brazilian Workshop on Social Network Analysis and Mining (BraSNAM), pages 61–74.

Rodrigues, F. F. (2025). A morfologia dos emojis: explorando a linguagem visual na comunicação digital em inglês. Trabalho de Conclusão de Curso. Universidade Estadual do Piauí.

Salles, I., Vargas, F., and Benevenuto, F. (2025). Hatebrxplain: A benchmark dataset with human-annotated rationales for explainable hate speech detection in brazilian portuguese. In Proceedings of the International Conference on Computational Linguistics (COLING), pages 6659–6669.

Silva, S. C. and Serapião, A. B. (2018). Detecção de discurso de ódio em português usando cnn combinada a vetores de palavras. In Symposium on Knowledge Discovery, Mining and Learning (KDMiLe), pages 1–8.

Trajano, D., Bordini, R. H., and Vieira, R. (2024). Olid-br: offensive language identification dataset for brazilian portuguese. Language Resources and Evaluation, 58(4):1263–1289.