A Search Strategy for Social Resource in Decentralized Social Networks

Peer-to-Peer (P2P) architecture is an appealing alternative to centralized architecture, especially for large-scale distributed internet applications. Owing to the similarity between human social networks and P2P social networks, where peer nodes are considered as persons, connections between peer nodes are considered as relationship, human social theories are used to construct P2P social network aiming to improve the performance and efficiency of resource discovery. In this paper, we propose a novel multi-function resource search strategy (MFRS) which can resolve complex queries for social resource discovery in P2P social networks. Firstly, each node in the P2P social networks builds its own knowledge base to facilitate the future search. During the search process, nodes in the network spontaneously form interest groups and discover resources in collaboration with each other. Furthermore, during the search period, peer nodes utilize a neighbour recommending algorithm and an adaptive routing algorithm to search resources more efficiently. Finally, the proposed strategy has been evaluated in a dynamic network. The experimental results validate the effectiveness of MFRS against existing methods.