Modeling influence on posting engagement in online social networks: Beyond neighborhood effects

Abstract

In many online social networks (OSN), Member A can observe the online participation of Member B and thus may be exposed to a shift in the online participation of B. Such exposure is often modeled by a neighbor-to-neighbor spreading process that can lead to local influence. In other words, exposure to participation shifts of Neighbor B can exert local influence on A’s posting engagement. Since many posts nowadays are public rather than private, exposure to participation shifts of non-neighbors outside A’s neighborhood can exert global influence on A’s posting engagement. We model influence spread in an OSN by differentiating between exposure to participation shifts of local and global influence sources. The results of applying the proposed model to different temporal dynamic OSNs show evidence of both local and global influence on posting engagement of members. This consideration of global influence in OSNs, beyond neighborhood effects, extends past research that focused mainly on local influence and provides new insights about influence sources in general and viral marketing in particular.