The 2018 Brazilian Presidential Run-Off: A Complex Network Analysis Approach Using Twitter Data

The political role of social networks has increased significantly in recent years. It has shifted from simple data analytics, employed to understand characteristics and desires of social groups, to a crucial digital platform for political campaigns and social influence. Here, we present an exploratory study of social data gathered from Twitter during the second round of the 2018 Brazilian Presidential Election. The study used complex network analysis to identify hidden communication patterns, important features, and key actors within Twitter data. The results were used to better understand how political polarization is embedded into Twitter during presidential campaigns. Moreover, the model also provided a straightforward manner to scrutinize the data and infer hypotheses from the obtained patterns. In summary, the number of communities of the network revealed an unbalanced division between the candidates and a lack of moderate behaviors.