Research On Adaptive Information Retrieval System Based On Agent

The quantity of Web information grows exponentially.How to quickly find user's interesting information is the major problem.This thesis designs and implements an adaptive information retrieval system, studying the core modules of search result clustering and user feedback.First of all,the paper presents the concept of information retrieval and meta search engine, discussing the intelligent Agent,search result clustering, and user feedback technologies.Then it focuses on technologies of search result clustering and user feedback,puts forward.a search result clustering algorithm based on tolerance class and improves the old linear reorder method considering the order quality of the result.Besides,the paper designs the meta search model based on intelligent Agent,improving the retrieval efficiency.Finally,it realizes the information retrieval system prototype,including the two modules.Simulation results show that the search clustering algorithm proposed can solve the problem of single result display,effectively balances the quality and time of clustering,and the improved user feedback method advances the order quality to a certain extent.