A novel social search model based on trust and popularity

An important function of online social networks (OSNs) is to help users find or add friends with common interests. In order to improve the quality of friends-searching results, a model for ranking these results is proposed in this paper with the introduction of two novel concepts named trust and popularity along with their calculation algorithm. The improved shortest path algorithm with the introduction of a threshold is used for computing the trust value in efficiency, and the page rank algorithm is used for solving the popularity value iteratively. Then the combination algorithm for these two values is performed to generate the final searching output. Experimental results show that this model can provide more satisfactory searching results for users, and provides wonderful supports for friend searching and friend recommendation in OSNs. This model has been deployed in "Scene Life", a mobile social networking system of Nokia Research Center Beijing.