P2P信息检索系统的查询结果排序与合并策略  

A Strategy of Query Result Ranking and Merging for P2P Information Retrieval Systems

作　　者：凌波[1] 周水庚[2] 周傲英[2] 

机构地区：[1]中国浦东干部学院信息技术部,上海201204 [2]复旦大学计算机科学与工程系,上海200433

出　　处：《计算机学报》2007年第3期405-414,共10页Chinese Journal of Computers

基　　金：国家自然科学基金(60373019;60573183和90612007)资助

摘　　要：基于P2P信息检索系统的特性,提出了一种完全分布式的查询结果排序与合并策略.首先分析当前P2P信息检索系统查询结果排序和合并问题的根源;接着提出一种完全分布式的查询结果排序与合并策略,包括元数据管理策略、查询结果的排序与合并的实现;然后用详细的实验证明了该策略的有效性.This paper proposes a fully distributed strategy to rank and merge the results retrieved from different peers in P2P information retrieval systems. First, the challenges of query answering in P2P IR systems are investigated and the issue of retrieved results ranking and merging is identified. Second, a fully distributed strategy is developed and its implementing issues are addressed. Finally, an extensive experimental study is condu.cted and the results verify the effectiveness of this proposed strategy.

关 键 词：P2P 信息检索 查询结果排序与合并 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]