Efficient Influential Individuals Discovery on Service-Oriented Social Networks: A Community-Based Approach

With the rapid development of Internet and mobile Internet, service-oriented social networks gain increasing popularity. Discovering a small subset of influential individuals on service-oriented social networks is beneficial for both users and service providers. This issue is formally referred to the influence maximization problem. In this paper, through exploiting the community structures of social networks, we propose two novel community-based approximation algorithms BCAA and ICAA, which have high performance guarantee as well as high efficiency, to address the influence maximization problem. Both BCAA and ICAA discover influential individuals within each individual community rather than the entire network. We further provide performance guarantee analysis of BCAA and ICAA. Finally, extensive experiments are conducted to demonstrate the efficiency and effectiveness of the proposed algorithms.