Clustering peers based on contents for efficient similarity search

  • Authors:
  • Yanfeng Shu;Bei Yu

  • Affiliations:
  • School of Computing, National University of Singapore, Singapore;School of Computing, National University of Singapore, Singapore

  • Venue:
  • DASFAA'06 Proceedings of the 11th international conference on Database Systems for Advanced Applications
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Similarity search is becoming a norm in most real-life applications such as digital asset management systems. In such systems, users typically want to retrieve documents or objects similar to terms specified in the query or query examples. In this paper, we present a system for supporting similarity search in P2P networks that retains many desirable properties of existing P2P systems. To support efficient search, peers are formed into clusters based on their contents and clusters are organized as a structured overlay. Optimizations are employed to improve search performance. The experimental results confirm the effectiveness of our proposed system architecture.