Characterizing Polarization in the 2018 and 2022 Brazilian Elections: An Analysis of Reddit Discussions with a Regression Model for Stance Detection
DOI:
https://doi.org/10.5753/reic.2025.6597Keywords:
Political Polarization, Reddit, Brazilian Elections, Stance Detection, Discussion Analysis, Social Media, Machine LearningAbstract
The polarization in political discussions on Reddit during the Brazilian presidential elections of 2018 and 2022 was investigated using a methodology based on a machine learning model for user stance detection in discussion trees. The findings reveal an increase in polarization and a decline in diverse and moderate debates over the analyzed periods. The use of open-source models enabled the analysis of 27,358 discussion trees comprising 204,296 comments, wich outperformed LLM GPT-4 zero-shot for the same task. The proposed polarization metrics demonstrated high correlation and consistency, validating the methodology's effectiveness. Qualitative analysis supported the quantitative findings, highlighting the formation of echo chambers and the prevalence of unilateral discussions in 2022 compared to 2018.
Downloads
References
Achiam, J., Adler, S., Agarwal, S., Ahmad, L., Akkaya, I., Aleman, F. L., Almeida, D., Altenschmidt, J., Altman, S., Anadkat, S., et al. (2023). Gpt-4 technical report. arXiv preprint arXiv:2303.08774.
Agrawal, R., Rajagopalan, S., Srikant, R., and Xu, Y. (2003). Mining newsgroups using networks arising from social behavior. In Proceedings of the 12th international conference on World Wide Web, pages 529–535.
ALDayel, A. and Magdy, W. (2021). Stance detection on social media: State of the art and trends. Information Processing & Management, 58(4):102597. DOI: 10.1016/j.ipm.2021.102597.
Almeida, R. J. A. (2018). Leia - léxico para inferência adaptada. [link].
Alsinet, T., Argelich, J., Béjar, R., and Martínez, S. (2021a). Measuring polarization in online debates. Applied Sciences, 11(24):11879.
Alsinet, T., Argelich, J., Béjar, R., and Martínez, S. (2021b). Measuring polarization in online debates. Applied Sciences, 11(24):11879.
Britt, R. K., Franco, C. L., and Jones, N. (2023). Trends and challenges within reddit and health communication research: A systematic review. Communication and the Public, 8(4):402–417.
Brum, P., Cândido Teixeira, M., Vimieiro, R., Araújo, E., Meira Jr, W., and Lobo Pappa, G. (2022). Political polarization on twitter during the covid-19 pandemic: a case study in brazil. Social Network Analysis and Mining, 12(1):140.
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 Proc. of the SocInfo.
Corrêa, E. V. B. (2023). Redes sociais, ódio e polarização política: A psicodinâmica da guerra civil digital brasileira. POLÍTICA EM FOCO: O melhor embate é o debate–Vol. 3, page 17.
Cunha, G. F. and da Silva, A. P. C. (2024). Caracterizando polarização em redes sociais: Um estudo de caso das discussões no reddit sobre as eleições brasileiras de 2018 e 2022. In Proceedings of the 30th Brazilian Symposium on Multimedia and the Web (WebMedia 2024). Sociedade Brasileira de Computação-SBC.
Dahlberg, L. (2007). Rethinking the fragmentation of the cyberpublic: from consensus to contestation. New media & society, 9(5):827–847.
de Arruda, H. F., Cardoso, F. M., de Arruda, G. F., Hernández, A. R., da Fontoura Costa, L., and Moreno, Y. (2022). Modelling how social network algorithms can influence opinion polarization. Information Sciences, 588:265–278.
Franco, A. B. and Pound, N. (2022). The foundations of bolsonaro’s support: Exploring the psychological underpinnings of political polarization in brazil. Journal of Community & Applied Social Psychology, 32(5):846–859.
Garimella, K., Morales, G. D. F., Gionis, A., and Mathioudakis, M. (2018). Quantifying controversy on social media. ACM Transactions on Social Computing, 1(1):1–27.
Guillou, P. (2021). Portuguese bert base cased qa (question answering), finetuned on squad v1.1.
Guimaraes, S., Silva, M., Caetano, J., Araújo, M., Santos, J., Reis, J. C., Silva, A. P., Benevenuto, F., and Almeida, J. M. (2022). Análise de propagandas eleitorais antecipadas no twitter. In Anais do BrasNAM.
Heltzel, G. and Laurin, K. (2020). Polarization in america: Two possible futures. Current opinion in behavioral sciences, 34:179–184.
Hutto, C. and Gilbert, E. (2014). Vader: A parsimonious rule-based model for sentiment analysis of social media text. In Proceedings of the international AAAI conference on web and social media, volume 8, pages 216–225.
Karypis, G. (1997). Metis: Unstructured graph partitioning and sparse matrix ordering system. Technical report.
Kim, Y., Hsu, S.-H., and de Zúñiga, H. G. (2013). Influence of social media use on discussion network heterogeneity and civic engagement: The moderating role of personality traits. Journal of communication, 63(3):498–516.
Liang, B., Li, A., Zhao, J., Gui, L., Yang, M., Yu, Y., Wong, K.-F., and Xu, R. (2024). Multi-modal stance detection: New datasets and model. arXiv preprint arXiv:2402.14298.
Lima, H. F. d. (2021). Polarização política afetiva: São os valores humanos e os traços de personalidade uma explicação?
Mutz, D. C. (2002). The consequences of cross-cutting networks for political participation. American journal of political science, pages 838–855.
Oliveira, D. F. N. (2022). Interpretabilidade de modelos de aprendizado profundo aplicados ao diagnóstico e prognóstico não supervisionado de falhas. PhD thesis, Universidade de São Paulo.
Ortellado, P., Ribeiro, M. M., and Zeine, L. (2022). Existe polarização política no brasil? análise das evidências em duas séries de pesquisas de opinião. Opinião Pública, 28:62–91.
Pereira, C., Pavan, M., Yoon, S., Ramos, R., Costa, P., Cavalheiro, L., and Paraboni, I. (2023). Ustancebr: a multimodal language resource for stance prediction. arXiv preprint arXiv:2312.06374.
Recuero, R., Soares, F. B., and Gruzd, A. (2020). Hyper-partisanship, disinformation and political conversations on twitter: The brazilian presidential election of 2018. In Proceedings of the international AAAI conference on Web and social media, volume 14, pages 569–578.
Rossini, P. (2020). Beyond toxicity in the online public sphere: understanding incivility in online political talk. A research agenda for digital politics, pages 160–170.
Ruediger, M. A. et al. (2014). Redes sociais retratam eleição mais polarizada da história recente do brasil.
Santos, P. D. and Goya, D. H. (2021). Automatic twitter stance detection on politically controversial issues: A study on covid-19’s cpi. In Anais do XVIII Encontro Nacional de Inteligência Artificial e Computacional, pages 524–535. SBC.
Saveski, M. (2020). Polarization and toxicity in political discourse online.
Silva, A. G. (2023). O insaciável espírito da época: ensaios de psicologia analítica e política, de: Humbertho oliveira, roque tadeu gui e rubens bragarnich. editora vozes, 2021. Self-Revista do Instituto Junguiano de São Paulo, 8:e007–e007.
Souza, F., Nogueira, R., and Lotufo, R. (2020a). BERTimbau: pretrained BERT models for Brazilian Portuguese. In 9th Brazilian Conference on Intelligent Systems, BRACIS, Rio Grande do Sul, Brazil, October 20-23 (to appear).
Souza, F., Nogueira, R., and Lotufo, R. (2020b). Bertimbau: pretrained bert models for brazilian portuguese. In Intelligent Systems: 9th Brazilian Conference, BRACIS 2020, Rio Grande, Brazil, October 20–23, 2020, Proceedings, Part I 9, pages 403–417. Springer.
Vicario, M. D., Quattrociocchi, W., Scala, A., and Zollo, F. (2019). Polarization and fake news: Early warning of potential misinformation targets. ACM Transactions on the Web (TWEB), 13(2):1–22.
Zhang, B., Ding, D., and Jing, L. (2022). How would stance detection techniques evolve after the launch of chatgpt? arXiv preprint arXiv:2212.14548.
Zhang, B., Fu, X., Ding, D., Huang, H., Li, Y., and Jing, L. (2023). Investigating chain-of-thought with chatgpt for stance detection on social media. arXiv preprint arXiv:2304.03087.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 The authors
This work is licensed under a Creative Commons Attribution 4.0 International License.
