Analysis of Online Electoral Advertising in 2022 Brazilian Elections Using Topic Modeling
摘要
This chapter presents the use of topic modeling to analyze advertisements boosted during the 2022 Brazilian Elections. Our goal was to investigate how the names of Luiz Inácio Lula da Silva and Jair Bolsonaro were used by candidates who were not running for the presidential position. The results showed that, in spite of the strong polarized context, the majority of the content was about the same general theme. We could observe the polarization on less than 10% of the ads analyzed.