Finding and evaluating the community structure in semantic peer-to-peer overlay networks

Abstract

Recently much attention has been paid to semantic overlay networks for information retrieval in large scale peer-to-peer networks, and much research work on semantic overlay protocols and searching algorithms has been done and the results indicate that semantic overlay is efficient for content searching in peer-to-peer networks. However, very limited work has been done to analyze and evaluate the characteristics of semantic overlay networks. In this paper we identify a natural property of semantic overlay networks, the community structure. We propose a mathematical model to evaluate the property of community structure of semantic P2P overlay networks. A heuristic algorithm is designed to optimize the community structure. Using the evaluation model we compare the SemreX semantic overlay with the Gnutella network. Results demonstrate that a SemreX overlay network has the distinctive community structure feature, while a Gnutella-like network does not. We also simulate a simple flooding protocol in both overlays to show that the overlay with community structure is more efficient for content searching.