Context-Aware Influence Diffusion in Online Social Networks

Abstract

In recent years, online social network becomes an important channel for people to communicate and spread innovations. Most studies reveal that the diffusion of influence messages significantly relies on the network topological structure. However, an individual context, referring to a set of beliefs towards various topics based on past experiences, impacts the influence adoption to a large extent. Moreover, the broadcasting approaches from various channels, e.g., advertisements from TV, policies deployed by a country, a piece of breaking news, famous scandals, etc., also drive the network evolutionary pattern. In this study, we model the influence diffusion in online social networks by considering individuals’ contexts and compare the influence propagation patterns under different scenarios. The results show that context-aware influence diffusion turns out to be an experienced model, where beliefs formed through users’ past experiences affect the adoption of influences.