Online information operations (IOs) refer to organized attempts to tamper with the regular flow of information and to influence public opinion. Coordinated online behavior is a tactic frequently used by IO perpetrators to boost the spread and outreach of their messages. However, the exploitation of coordinated behavior within large-scale IOs is still largely unexplored. Here, we build a novel dataset comprising around 624K users and 4M tweets to study how online coordination was used in two recent IOs carried out on Twitter. We investigate the interplay between coordinated behavior and IOs with state-of-the-art network science and coordination detection methods, providing evidence that the perpetrators of both IOs were indeed strongly coordinated. Furthermore, we propose quantitative indicators and analyses to study the different patterns of coordination, uncovering a malicious group of users that managed to hold a central position in the discussion network, and others who remained at the periphery of the network, with limited interactions with genuine users. The nuanced results enabled by our analysis provide insights into the strategies, development, and effectiveness of the IOs. Overall, our results demonstrate that the analysis of coordinated behavior in IOs can contribute to safeguarding the integrity of online platforms.
Coordinated Behavior in Information Operations on Twitter
Cima, Lorenzo;Mannocci, Lorenzo;Avvenuti, Marco;
2024-01-01
Abstract
Online information operations (IOs) refer to organized attempts to tamper with the regular flow of information and to influence public opinion. Coordinated online behavior is a tactic frequently used by IO perpetrators to boost the spread and outreach of their messages. However, the exploitation of coordinated behavior within large-scale IOs is still largely unexplored. Here, we build a novel dataset comprising around 624K users and 4M tweets to study how online coordination was used in two recent IOs carried out on Twitter. We investigate the interplay between coordinated behavior and IOs with state-of-the-art network science and coordination detection methods, providing evidence that the perpetrators of both IOs were indeed strongly coordinated. Furthermore, we propose quantitative indicators and analyses to study the different patterns of coordination, uncovering a malicious group of users that managed to hold a central position in the discussion network, and others who remained at the periphery of the network, with limited interactions with genuine users. The nuanced results enabled by our analysis provide insights into the strategies, development, and effectiveness of the IOs. Overall, our results demonstrate that the analysis of coordinated behavior in IOs can contribute to safeguarding the integrity of online platforms.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.