Exploring populism in Jair Bolsonaro's speeches: a statistical study

Authors

DOI:

https://doi.org/10.51359/2317-0115.2024.265421

Keywords:

Jair Bolsonaro, Machine Learning, Support Vector Classification algorithm

Abstract

This study seeks to provide an initial analysis using a corpus derived from the speeches delivered during the first year of former President Jair Bolsonaro's administration. Before assessing the performance of several widely-used machine learning algorithms, basic pre-processing techniques were applied to the data. The findings revealed that the Support Vector Classification
algorithm achieved the highest performance, with an accuracy rate of 86%. Furthermore, the study included an analysis based on bigrams to distinguish and compare the linguistic patterns between speeches classified as populist and those considered non-populist. These initial results offer valuable insights into the rhetorical strategies employed during Bolsonaro's early presidency.

References

Canto, I. Teorias Rivais sobre a desconsolidação da democracia. Trabalho de Conclusão de Curso (Bacharelado em Ciência Política), Universidade Federal de Pernambuco. Recife, p. 56. 2022

Engesser, S.; Ernst, N.; Esser, F.; Büchel, F. (2017) Populism and social media: how politicians spread a fragmented ideology, Information, Communication & Society, 20:8, 1109-1126, DOI: 10.1080/1369118X.2016.1207697

Joshi, Y. (2020). A quick guide to text cleaning using the nltk library. Analytics Vidhya. Available at: https://www.analyticsvidhya.com/blog/2020/11/text- cleaning-nltk-library/h2 2 .

Kirk, H., Rovira, C. (2017). What the (ideational) study of populism can teach us, and what it can’t. Swiss Political Science Review, 23(4).

LUST, Ellen; WALDNER, David. Theories of Democratic change. Retrived from https://pdf. usaid. gov/pdf_docs/PBAAD635. pdf, 2015.

Moffit, Benjamin. The Global Rise of Populism: Performance, Political Style, and Representation. Stanford University Press, 2016.

Nicolau, J. (2020). O Brasil dobrou a` direita: Uma radiografia da elei¸c˜ao de Bolsonaro em 2018. Rio de Janeiro: Zahar. 141 p.

Pérez-Liñán (2018). A Impeachment or backsliding? Threats to democracy in the twenty-first century, Revista Brasileira de Ciências Sociais 33;

Ricci, P., Izumi, M., Moreira, D. (2021). O populismo no Brasil (1985-2019): um velho conceito a partir de uma nova abordagem. Revista Brasileira de Ciências Sociais, 36(107), 1-22. Available at: https://doi.org/10.1590/3610707/2021. Acesso em: 20 ago.

Torfi, A., Shirvani, R. A., Keneshloo, Y., Tavvaf, N., Fox, E. (2020). Natural language processing advancements by deep learning: A survey. ArXiv, vol. abs/2003.01200.

Published

2025-03-20