Analyzing the Acceptance of the 2018 Brazilian Presidential Election’ Main Candidates Based on YouTube Comments

• Cristian A. Silva UFV
• Daniel M. Barbosa UFV

Resumo

Online Social Networks (OSNs) are becoming, in general, increasingly popular and used in a global scope, and consequently, the analysis of the information and the content generated by its users become more and more important in different contexts. According to this importance and to the current tendency of discussing topics related to politics in these environments, the objective of this work is to analyze the acceptance of the two candidates who disputed in the second shift of the Brazilian presidential election of 2018 (Fernando Haddad and Jair Messias Bolsonaro) since the beginning of the electoral campaign using a dataset of comments collected from YouTube. Our results demonstrate that, after the detection of the political positioning and the classification of the sentiment of the comments, by disregarding all the comments that are identically related to the two candidates, Bolsonaro obtained an acceptance by the users at least 16.1% and 15.4% higher than Haddad on both electoral shifts, considering or not the likes of the comments, respectively.