Modeling disinformation networks on Twitter: structure, behavior, and impact

Abstract

The influence and pervasiveness of misinformation on social media platforms such as Twitter have been well-documented in recent years. These platforms’ real-time, rapid-fire nature and the personalized, echo-chamber-like environments they foster, often inadvertently, assist in misinformation amplification. To better understand this situation and how to encourage safer and broader narratives, this paper presents a comparative study of the activity of 275 Twitter accounts tagged as disinformation sources and 275 accounts tagged as legitimate journalists over a 3.5-year period in the Spanish context. By employing various modeling techniques, we investigate the structural differences and behavioral patterns between the two groups. Our findings demonstrate that disinformation accounts exhibit a coordinated behavior, among other distinct characteristics, leading to more efficient (dis)information propagation. The implications of these findings for understanding the dynamics of disinformation networks and combating their impact are discussed.