forP2PInformationRetrieval.PDFVIP

International Symposium on Applications and the Internet Huffman-DHT: Index Structure Refinement Scheme for P2P Information Retrieval Hisashi Kurasawa Atsuhiro Takasu Jun Adachi The University of Tokyo National Institute of Informatics 2-1-2 Hitotsubashi, Chiyoda-ku, Tokyo, JAPAN 2-1-2 Hitotsubashi, Chiyoda-ku, Tokyo, JAPAN kurasawa@nii.ac.jp {takasu,adachi}@nii.ac.jp Abstract relevance between queries and documents [5]. Although the term information can be easily calculated in centralized Peer-to-Peer Information Retrieval (P2P IR) systems us- systems, it is difficult problem in P2P systems. Early P2P ing a distributed index on a distributed hash table (DHT) systems [2] flooded queries over P2P networks and gath- can make highly precise searches for documents relevant ered documents that contained the query terms. They did to a query. However, these systems require a heavy index not fully use the term information, and therefore, the gath- construction cost, and cause unfair index management costs ered documents contained less relevant documents. They due to the unbalanced term frequency distribution. We pro- also caused heavy network traffic loads. Recently, several pose a new node access scheme for P2P IR that we call distributed indexing methods for P2P IR systems have been