Modeling Online Social Networks that React to External Influencers
Conference
65th ISI World Statistics Congress 2025
Format: IPS Abstract - WSC 2025
Keywords: complex networks, social network analysis
Session: IPS 924 - When Machine Learning meets High Dimensional Networks
Tuesday 7 October 10:50 a.m. - 12:30 p.m. (Europe/Amsterdam)
Abstract
The influence of individuals via social media platforms results in real change to the behavior of those being influenced. An entire industry of “Influencers” has arisen because of their ability to affect consumer behavior. Yet there are many types of Influencers online. We consider what we name “Strategic Influencers”: individuals who attempt to change the attitudes and behavior of others on behalf of an external entity (e.g. a paid relationship with a company). Yet measuring the influence that Influencers have on the attitudes/opinions and behavior of others is difficult absent the strict controls of a laboratory experiment. This is due, in part, to the counterfactual of how the social network would have fluctuated if the influencer were not present. We develop a methodology for identifying and measuring influence online and use it to analyze the effects of influencers on social media platforms. We further use communication patterns to discriminate Strategic Influencers from others and thus measure the effects an external entity can bring to bear.