{"id":2236,"date":"2024-06-04T09:51:46","date_gmt":"2024-06-04T07:51:46","guid":{"rendered":"https:\/\/www.iit.cnr.it\/?post_type=news&p=2236"},"modified":"2024-06-04T09:51:47","modified_gmt":"2024-06-04T07:51:47","slug":"coordinated-behaviour-online-paper-pnas-cresci","status":"publish","type":"news","link":"https:\/\/www.iit.cnr.it\/en\/news\/coordinated-behaviour-online-paper-pnas-cresci\/","title":{"rendered":"Those coordinated groups that pollute the social media debate during political elections"},"content":{"rendered":"\n

Published in PNAS,\u00a0the Proceedings of National Academy of Sciences\u00a0\u2014 one of the world’s most-cited and prestigious scientific journals \u2014 a study led by Stefano Cresci<\/a><\/strong> (IIT-CNR) present a groundbreaking analysis of coordinated online behavior<\/strong>. The results of the study titled\u00a0“Temporal Dynamics of Coordinated Online Behavior: Stability, Archetypes, and Influence”<\/a>\u00a0shed new light on a fundamental dynamic of online interactions with practical implications for understanding opinion formation and for countering online manipulations.

Organized social media campaigns<\/strong> are among the most effective tools for shaping public opinion in various spheres of society, reaching an impressive number of people, citizens, and voters. This form of communication relies on a large number of accounts, often also including fake accounts such as trolls and bots, working in a coordinated manner to pursue the goals of the campaign. Current methods used to detect coordinated behavior on social media perform static analyses<\/strong>, while the new study by Cresci and colleagues relies on a multi-layer temporal network and dynamic community detection algorithms<\/strong> to identify groups of users who exhibit coordinated behaviors in time.

\"\"“We focused on the temporal dynamics of coordinated campaigns on Twitter during two recent major elections, the 2019 UK general election and the US 2020 presidential election, and this new approach gave us very interesting results,”<\/em>\u00a0explains Serena Tardelli<\/strong>, first author of the study.\u00a0“The analysis revealed unknown patterns of opinion formation, such as those that led some users to move from one side of the political spectrum to the other in the weeks preceding the vote. For example, we saw that in the case of the UK 2019 election, Labour communities gradually attracted users from Conservative communities as the debate unfolded.”<\/em>

“Another interesting finding of this study,”\u00a0<\/em>adds Stefano Cresci,\u00a0“is the\u00a0identification of some archetypal types of users that exhibit very\u00a0different behaviors: there are, for example, the undecided,\u00a0who do not attach to any group or political party; the loyalists who, on the contrary, never move from their community of reference; and those who become persuaded, who move from one community to another without ever leaving it.\u00a0Thus, our methodology\u00a0makes it possible to study the evolution of online debates much better than before, opening up unprecedented scenarios such as the possibility to anticipate trends of influence and persuasion in online discussions.”<\/em>

\"\"Stefano Cresci, Researcher at the Institute of Informatics and Telematics of the Italian National Research Council in Pisa, has been studying the issue of coordinated online behavior for years<\/strong>. His research involves developing new methods to identify online misbehaviors and to allow platforms to take countermeasures thus improving the trustworthiness and inclusivity of the online environments. On these topics, a few months ago Cresci also won a prestigious ERC Starting Grant<\/strong>, a 1.5 million euro prize to develop a research project \u2014 DEDUCE <\/strong>\u2014 dedicated to content moderation on social media.

In addition to Tardelli and Cresci, the study is authored by Leonardo Nizzoli<\/strong> and Maurizio Tesconi <\/strong>(IIT-CNR), Mauro Conti<\/strong> and Giovanni Da San Martino <\/strong>(University of Padua), and Preslav Nakov<\/strong> (MBZUAI).<\/p>\n","protected":false},"featured_media":2227,"template":"","categories":[],"tags":[389,25,390],"class_list":["post-2236","news","type-news","status-publish","has-post-thumbnail","hentry","tag-pnas-2","tag-research","tag-social-media-2"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.iit.cnr.it\/en\/wp-json\/wp\/v2\/news\/2236","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.iit.cnr.it\/en\/wp-json\/wp\/v2\/news"}],"about":[{"href":"https:\/\/www.iit.cnr.it\/en\/wp-json\/wp\/v2\/types\/news"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.iit.cnr.it\/en\/wp-json\/wp\/v2\/media\/2227"}],"wp:attachment":[{"href":"https:\/\/www.iit.cnr.it\/en\/wp-json\/wp\/v2\/media?parent=2236"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.iit.cnr.it\/en\/wp-json\/wp\/v2\/categories?post=2236"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.iit.cnr.it\/en\/wp-json\/wp\/v2\/tags?post=2236"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}