On Charactering of Word-of-Mouth Propagation in Heterogeneous Online Social Networks

Abstract

Recent years have witnessed the explosive growth of research on online social networks (OSNs), which provide a perfect platform for observing the Word-of-Mouth (WOM) propagation. However, recent research regarded OSNs as homogeneous networks, while the real-world OSNs are heterogeneous. This paper considered the heterogeneity of OSNs, proposed a WOM propagation model called HOSN to simulate WOM propagation in different types of nodes in OSNs. Numerical simulation experiments based on agents have been extensively conducted, which show the mechanism of WOM propagation in heterogeneous online social networks. It can be seen from the experimental results that along with the WOM spreading and time passing, more and more users in OSNs review and know the WOM information. Moreover, from the network structure and user behavior perspective, some interesting findings are got. First, the number of initial known nodes has a certain impact on the WOM propagation scope and speed. Then, the number of neighbors has a great impact on the WOM propagation process and results. Finally, percentage of different user type is the decisive factor for WOM propagation either in final propagated number or propagation speed. Findings in this paper attempt to complement the theoretical framework of WOM propagation and complex networks, and thus to further promote the practice of WOM control in OSNs