Document Type : Original Article
Department of Computer Science, IPM, Tehran, Iran
2 Department of Finance, Iranian E-Institute of Higher Education, Tehran, Iran
Social networks are becoming an easy to use platform for viral marketing that are much more powerful and fast in propagating considered information in different topics. To this end, identification of influential users in social networks plays a crucial role in a successful viral marketing. Most of existing influential maximization methods are based on structural properties of networks. Whereas there are more personal information such as users’ interests and friends’ common interests that affects the behavior of users in confronting the shared massages. This manuscript proposes a novel method to identify the influential users for marketing in social networks based on their specific interests and power of influencing on their neighbors. We claim that not any hub node can be chosen as the influential spreader in the considered marketing contents and the influential users should be chosen based on their interests topics obtained from their historical activities .We propose a new method to identify the most influential users. In the proposed method, the extent of interest of the influential nodes and their neighbors are considered and the SIR spreading model is used to investigate the spreading process. Experimental results on six real social networks reveal effectiveness of the proposed method as compared to the existing methods based on centrality measures.