Bayesian method based trusted overlay for information retrieval over networks

Abstract

Peer-to-peer (P2P) overlay networks provide a new way to retrieve information over networks. How to assurance the reliability of the resource is the crucial issue of security. This paper proposes a trustworthy overlay based on the small world phenomenon that facilitates efficient search for information retrieval with security assurance in unstructured P2P systems. Each node maintains a number of short-range links to the trusted other nodes, together with a small collection of long-range links that help increasing recall rate of information retrieval, and the trust degree of each node is evaluated by Bayesian method. Simulation test shows that the proposed model can not only increase the ratio of resource discovery, improve the interaction performance of the entire network, but also assurance the reliability of resource selection. © Springer-Verlag Berlin Heidelberg 2007.