Hou Xiang-song①; Cao Yuan-da②; Guan Zhi-tao③; Zhang Yu①
①School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China; ②School of Software, Beijing Institute of Technology, Beijing 100081, China;③Department of Computer, North China Electric Power University, Beijing 112206, China
Abstract：Peer-to-Peer (P2P) overlays are appealing, since they can aggregate resources of end systems without relying on sophisticated infrastructures. Unfortunately current peer-to-peer systems either offer exact keyword match or provide inefficient text search methods through centralized indexing or flooding. In this paper, a semantic search system is proposed for structured P2P overlays without relying on message flooding. LSH is used to map semantically related text vector to nearly node, and a mechanism is carefully designed to cope with load balancing. Experimental results show that this is a steady system with high recall, good load balance.
2009, Vol. 31 
