P2P网络中一种新型的基于DHT的多关键字信息检索模型  

A novel DHT-based multiple keywords information search model in P2P networks

作　　者：周健[1] 洪佩琳[1] 李津生[1] 

机构地区：[1]中国科学技术大学电子工程与信息科学系信息网络实验室,安徽合肥230027

出　　处：《电路与系统学报》2008年第3期23-29,67,共8页Journal of Circuits and Systems

基　　金：国家自然科学基金资助项目(90104011)

摘　　要：在现有的P2P系统应用中,信息的检索是一项非常重要的内容,但现有的检索机制不够高效。本文针对这个问题,在现有的结构化DHT模型基础上,提出一种新型的多关键字检索模型。在一个P2P节点上存储包含一个资源内容的多个关键字的信息,使用带有权重的多关键字向量代表资源信息,并把请求也表示成带有权重的多关键字向量,对请求和资源信息的向量进行匹配,选取符合要求的信息。仿真和分析表明此方案可以显著改善搜索结果的精确性和大大减少搜索所需的寻路跳数。Information search is a crucial part in peer-to-peer application systems. But existing search mechanisms are obsessed with low efficiency. We propose a multiple keywords search model based on DHTs to solve the problem. Information of multiple keywords associated with the same resource is stored in one P2P node, and weighted vectors are used to denote information of the resource. Search queries are also expressed as weighted vectors. Then desired information can be located by matching vectors of queries and those of information. The simulation results show it can improve search accuracy significantly and reduce routing hops effectively.

关 键 词：P2P DHT 多关键字 信息检索 

分 类 号：TN919.2[电子电信—通信与信息系统]