• comparative analysis
• conspiracy propagators
• profile characterization
• social media analysis

The rise of social media has offered a fast and easy way for the propagation of conspiracy theories and other types of disinformation. Events in recent years such as the COVID-19 pandemic have also helped this phenomenon grow more and more. Despite the research attention that has received, conspiracy remains an open problem that has the potential to cause harm both to the individual and the society as a whole. In this thesis we focus on the characterization of conspiracy user profiles on Twitter, one of the most used social networks to spread conspiracy theories, through a comparative analysis with a random user sample. For the comparison to be meaningful, random users must be randomly selected from those who talk about the same topics as the conspirators and whose accounts were created in the same years. To this end, first, we develop a strategy to identify a quite large sample of users who are highly likely to be conspirators. Then, we collect Twitter timelines and profile information of the selected group of conspirators through Twitter API and we mainly explore which topics they focus on the most and how their Twitter accounts are distributed by year of creation. Once random users are chosen, we perform a comparative analysis over various profile characteristics between the two samples. The results show statistically significant differences between conspiracy users and random users and highlight the distinctive and peculiar traits of conspirators that emerged, by considering the productivity in terms of tweets, the social networks and the estimation of demographic data.