The effect of political polarization on social distance stances in the Brazilian COVID-19 scenario

The COVID-19 pandemic changed the routine and concerns of people around the world since 2020. The alarming contagious rate and the lack of treatment or vaccine evoked different reactions to controlling and mitigating the virus's contagious. In this paper, we developed a case study on the Brazilian COVID scenario, investigating the influence of the political polarization in the pro/against stances of social isolation, represented in Twitter by two groups referred to as the Cloroquiners and Quarenteners. We analyzed these groups according to multiple dimensions: a) concerns expressed by each group and main arguments representing each stance; b) techniques to automatically infer from users political orientation, c) network analysis and community detection to characterize their behavior as a social network group and d) analysis of linguistic characteristics to identify psychological aspects. We propose combining two topic modeling techniques, LDA and BERTopics, to understand each stance's concerns in different granularity levels. Our main findings confirm that Cloroquiners are right-wing partisans, whereas Quarenteners are more related to the left-wing. Cloroquiners and Quarenteners' political polarization influences the arguments of economy and life and a stronger support/opposition to the president. As a group, the network of Cloroquiners is more closed and connected, and Quarenteners have a more diverse political engagement with a community of users polarized only with left-wing politicians and his supporters. In terms of psychological aspects, polarized groups come together on cognitive issues and negative emotions.

[1]  Karin Becker,et al.  A framework to analyze the emotional reactions to mass violent events on Twitter and influential factors , 2020, Inf. Process. Manag..

[2]  Yael V. Hochberg,et al.  Risk Perception Through the Lens of Politics in the Time of the Covid-19 Pandemic , 2020, SSRN Electronic Journal.

[3]  Andrea Esuli,et al.  SentiWordNet 3.0: An Enhanced Lexical Resource for Sentiment Analysis and Opinion Mining , 2010, LREC.

[4]  C. Makridis,et al.  The Real Cost of Political Polarization: Evidence from the COVID-19 Pandemic , 2020, SSRN Electronic Journal.

[5]  Quoc V. Le,et al.  Distributed Representations of Sentences and Documents , 2014, ICML.

[6]  Mai ElSherief,et al.  #NotOkay: Understanding Gender-Based Violence in Social Media , 2017, ICWSM.

[7]  Karin Becker,et al.  Caracterização e Comparação de Campanhas Promovendo o Outubro Rosa e o Novembro Azul no Twitter , 2018, SBBD Companion.

[8]  Kristina Lerman,et al.  Political Partisanship and Anti-Science Attitudes in Online Discussions about Covid-19 , 2020, ArXiv.

[9]  Hyunghoon Cho,et al.  Density-Preserving Data Visualization Unveils Dynamic Patterns of Single-Cell Transcriptomic Variability , 2020, bioRxiv.

[10]  Javam C. Machado,et al.  An Analysis of Machine Learning Techniques to Prioritize Customer Service Through Social Networks , 2018, J. Inf. Data Manag..

[11]  Catherine Ordun,et al.  Exploratory Analysis of Covid-19 Tweets using Topic Modeling, UMAP, and DiGraphs , 2020, ArXiv.

[12]  Ben Shneiderman,et al.  Social network analysis: Measuring, mapping, and modeling collections of connections , 2020, Analyzing Social Media Networks with NodeXL.

[13]  Daniel Barbará,et al.  Topic Significance Ranking of LDA Generative Models , 2009, ECML/PKDD.

[14]  Lyle H. Ungar,et al.  Beyond Binary Labels: Political Ideology Prediction of Twitter Users , 2017, ACL.

[15]  J. Pennebaker,et al.  The Psychological Meaning of Words: LIWC and Computerized Text Analysis Methods , 2010 .

[16]  Munmun De Choudhury,et al.  Social Media Participation in an Activist Movement for Racial Equality , 2016, ICWSM.

[17]  Michael Röder,et al.  Exploring the Space of Topic Coherence Measures , 2015, WSDM.

[18]  Soojong Kim,et al.  Political partisanship influences behavioral responses to governors’ recommendations for COVID-19 prevention in the United States , 2020, Proceedings of the National Academy of Sciences.

[20]  Kurt Hornik,et al.  topicmodels : An R Package for Fitting Topic Models , 2016 .

[21]  Cindy K. Chung,et al.  Winning words: Individual differences in linguistic style among U.S. presidential and vice presidential candidates , 2007 .

[22]  Saif Mohammad,et al.  CROWDSOURCING A WORD–EMOTION ASSOCIATION LEXICON , 2013, Comput. Intell..

[23]  Ana L. C. Bazzan,et al.  I will be there for you: clique, character centrality, and community detection in Friends , 2020, Computational and Applied Mathematics.

[24]  Punam Bedi,et al.  Community detection in social networks , 2016, WIREs Data Mining Knowl. Discov..

[25]  Santo Fortunato,et al.  Community detection in graphs , 2009, ArXiv.

[26]  Igor Douven,et al.  Measuring coherence , 2006, Synthese.

[27]  Kristina Lerman,et al.  Political polarization drives online conversations about COVID‐19 in the United States , 2020, Human behavior and emerging technologies.

[28]  Leland McInnes,et al.  Accelerated Hierarchical Density Based Clustering , 2017, 2017 IEEE International Conference on Data Mining Workshops (ICDMW).

[29]  Jean-Loup Guillaume,et al.  Fast unfolding of communities in large networks , 2008, 0803.0476.

[30]  R. Recuero,et al.  Research note: Bolsonaro’s firehose: How Covid-19 disinformation on WhatsApp was used to fight a government political crisis in Brazil , 2021 .

[31]  Mohammad Al Hasan,et al.  Dynamic topic modeling of the COVID-19 Twitter narrative among U.S. governors and cabinet executives , 2020, ICWSM Workshops.

[32]  Cláudio de Souza Baptista,et al.  Using Open Data to Analyze Urban Mobility from Social Networks , 2017, J. Inf. Data Manag..

[33]  Jorge E. Camargo,et al.  Using machine learning and information visualisation for discovering latent topics in Twitter news , 2019, ArXiv.

[34]  James Zou,et al.  Analyzing Polarization in Social Media: Method and Application to Tweets on 21 Mass Shootings , 2019, NAACL.

[35]  Karin Becker,et al.  Exploring Deep Learning for the Analysis of Emotional Reactions to Terrorist Events on Twitter , 2019, J. Inf. Data Manag..

[36]  L. da F. Costa,et al.  Characterization of complex networks: A survey of measurements , 2005, cond-mat/0505185.

[37]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[38]  Roberto de Alencar Lotufo,et al.  BERTimbau: Pretrained BERT Models for Brazilian Portuguese , 2020, BRACIS.

[39]  Venkata Rama Kiran Garimella,et al.  A Long-Term Analysis of Polarization on Twitter , 2017, ICWSM.

[40]  Quarenteners vs. Cloroquiners: a framework to analyze the effect of political polarization on social distance stances , 2020 .

[41]  Jacek M. Zurada,et al.  Normalized Mutual Information Feature Selection , 2009, IEEE Transactions on Neural Networks.

[42]  Jacob Ratkiewicz,et al.  Political Polarization on Twitter , 2011, ICWSM.

[43]  Christopher D. Manning,et al.  Introduction to Information Retrieval , 2010, J. Assoc. Inf. Sci. Technol..

[44]  Sun Hyoung Kim,et al.  Political polarization on twitter: Implications for the use of social media in digital governments , 2016, Gov. Inf. Q..

[45]  Quarenteners vs. Chloroquiners: A framework to analyze how political polarization affects the behavior of groups , 2020, 2020 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT).

[46]  Guilherme Dal Bianco,et al.  Exploratory Analysis of Electronic Health Records using Topic Modeling , 2021, J. Inf. Data Manag..

[47]  Arthur Spirling,et al.  Text Preprocessing For Unsupervised Learning: Why It Matters, When It Misleads, And What To Do About It , 2017, Political Analysis.

[48]  Nicolás Ajzenman,et al.  More than Words: Leaders' Speech and Risky Behavior During a Pandemic , 2020, SSRN Electronic Journal.