ABSTRACT
Online social networks are at the limelight of the public debate, where antagonistic groups compete to impose conflicting narratives and polarize the discussions. This article proposes an approach for measuring network polarization and political sectarianism in Twitter based on user interaction networks. Centrality metrics identify a small group of influential users (polarizers and unpolarizers) who influence a larger group of users (polarizees and unpolarizees) according to their ideological stance (left, right, and undefined). This network polarization is computed by the Bayesian probability using typical actions such as following, tweeting, retweeting, and replying. The measurement of political sectarianism also uses Bayesian probability and words extracted from the tweets to quantify the intensity of othering, aversion, and moralization in the debate. We collected Twitter data from 33 conflicted political events in Brazil during 2020, strongly influenced by the COVID-19 pandemic. Based on our methodology and polarization score, our results reveal that the approach based on user interaction networks leads to an increasing understanding of polarized conflicts in Twitter. Also, a small number of polarizers is enough to represent the polarization and sectarianism of Twitter events. © 2014 IEEE.
ABSTRACT
As society becomes digitalized, online social networks tend to be primary places for debate but can turn into a battlefield for imposing conflicting narratives. Automating the identification of online conflicts is a challenge due to difficulties in defining antagonist communities and controversial discussions. Here, we propose a polarization approach for understanding Twitter conflicts in Brazil during the COVID-19 pandemic, where a small group of polarizers influences a larger group of polarizees according to their ideological leaning. Polarizers are automatically identified by centrality metrics in following, retweet, and reply networks and manually labeled as leftists, rightists, or undefined. We collected and analyzed the polarization of 21 potentially conflicted political events in Brazil. Our results show that polarizers adequately represent the polarization of events, the traditional media is giving way to a new breed of tweeters, and retweet and reply play different roles within a conflict that reflects their polarization level. © 2021 IEEE Computer Society. All rights reserved.
ABSTRACT
Concerns about the advances of the COVID-19 epidemic have sparked many debates around the world. One such discussion revolved around the use of the drug called chloroquine, initially thought to be effective in reducing the mortality rate of the infection. Particularly in Brazil, even after new studies pointed to the drug's ineffectiveness, the federal government kept the recommendation of this drug as an official treatment. The publication of an official authorization of the use of chloroquine on Twitter sparked an intense debate on social media with arguments against and in favor. This paper studies the dynamics of interactions among different user groups around this discussion, relying on network science and topic modeling analyses. Our results highlight two distinct behaviors in Twitter interaction networks, where retweets serve as positive reinforcements within information bubbles and replies act as a space of direct debate. Also, discussions are seeded by public figures, but regular users carry on the debate per se. The topic modeling analyses revealed three observable user groups in this debate: strong supporters of the Brazilian government, progressive opposition to this government, and moderate users that oppose to this specific topic but do not reject the government agenda as a whole. © 2021 IEEE Computer Society. All rights reserved.