Probabilistic reasoning system for social influence analysis in online social networks

Abstract

People interact with other people in their daily life, either for work or for personal reasons. These interactions are often complex. Thus, interactions that an individual has with other individuals, to some extent, influence the decisions they make. There have been many efforts to uncover, explore, and measure the concept of social influence. Thus, modeling influence is an open and challenging problem where most evaluation models focus on online social networks. However, they fail to characterize some properties of social influence. To address the limitations of the previous approaches, we propose a novel Probabilistic Reasoning system for social INfluence analysis (PRIN) to examine the social influence process and elucidate the factors that affect it in an attempt to explain this phenomenon. In this paper, we present a model that quantitatively measures social influence in online social networks. Experiments on a real social network such as Twitter demonstrate that the proposed model significantly outperforms traditional feature engineering-based approaches. This suggests the effectiveness of this novel model when modeling and predicting social influence.