TDA and the School of Communication will host “Analyzing Twitter Data: A Case Study of Presidential Impeachment in Brazil,” presented by three faculty members from Federal University of ABC: Denise Hideko Goya, PhD (computer science); Fabricio de Franca, PhD (computer science); and Claudio Penteado, PhD (political science).
About their talk: The impeachment of Dilma Rousseff in August 2016 was the result of a long process of political and economic events, which were followed by intense popular demonstrations in the streets as well as in online social networks. Since 2015, Brazilians have used the Twitter social network to express opinions and arguments for and against the impeachment, which generated a large amount of data publicly available for analyzing the network structures formed by user interactions.
Computational analysis of these data can identify influential individuals and the formation of groups of interactions and, also, it can reveal dominant topics allowing us to estimate which sentiment prevails. In this talk, we present a case study involving data collected from Twitter in April 2016 when the Brazilian National Congress voted to open the impeachment process against Rousseff. We explain the methods used to categorize users participating in the discussions and the techniques to analyze the content and the sentiment associated to the messages. We also talk about techniques for analyzing the network structures formed by the user interactions. Our methodological approach may be employed in the analysis of any large events, not only political ones.