Research And Implementation Of Enterprise Search Engine System Based On Lucene

At present,with the development of informatization，enterprise information isbecoming rich more and more, including not only a variety of database information, butalso a large number of text information. However, these text information are scattereddistribution and information organization to be diverse, are need to be uinfinedintegration Processing. The establishment of the enterprise information search system,optimized for enterprise information retrieval is becoming more and more important.The status and problems of enterprise search engine were analyzed, based on theLucene search engine system for the field of text documents was established, we alsoexpand the function of Lucene, focused on the sorting algorithms of Lucene and theimprovement of model, applying AHP it identifies appraisal indicators weight. In order tofacilitate users to browse, use the document search results KNN clustering algorithmauto-clustering, create a similar folder hierarchy is a good method. Multi-threadstechnique is used indistributed index management, which makes the system moreefficient and steady. In this paper,the system will processes targeted text extracting andindex fix-format converting on a sort of formats of documents such pdf、word、html、textand so on,it will also helps on the content indexing and storage in order to supportfull-text searching on various kinds of documents.At last by analyzing the case of Chongqing SME Infromation Resource library，this paper also describes the detail design and analysis on all parts of implementationmodules in enterprise search engine system and implements a enterprise search enginesystem by using SSH. Through a uniform interface, enterprise search engine select thequery keyword, then arrange the search results, finally return the results to users. Resultsshow that the system can meet the requirements of the users，it also indicated that thesystem was feasible and practical.