Resource Sharing Behavior in a Socialized Peer-to-Peer Internet Environment

Abstract

Retrieving resources which are interesting, useful, relevant, and valuable to users is always a challenge for researchers and vendors of Information Retrieval in a Peer-to-Peer (P2PIR) environment. Current research of document-based P2PIR focused mainly on the feasibility of extending IR operations and the concept of social network analysis (SNA) has been adopted as a methodology to model interrelationships among actors in ad hoc environments. The authors demonstrate the possibility to discover characteristics of users in a P2P environment by means of the Zipf long-tail distribution. It can efficiently categorize the shared information with small portions of its contents. The results indicate that for any document corpus shared by users in a P2P environment the long-tail distribution applies also. Since behavioral profiles of users can be systematically and efficiently discovered by analysis based on the long-tail distribution, the research result of this socialized approach to resource discovery in a P2P environment can provide a feasible and efficient solution.