The Design And Implementation Of Information Retrieval System Based On Enterprise Collaboration Information Network Platform

This article mainly introduces the realization of manufacture enterprise collaborative information network platform and the design and implementation of information retrieval sub system based on it to meet the needs of manufacture enterprises users. The manufacture enterprise collaborative information network platform provides a B/S mode information management way, and the retrieval sub system is mainly based on vector space model; it carries on analysis processing of user's inquiry requests, matches with the backstage documents collection about cooperative information, and returns the retrieval result to users. The goal of the system is to feed back the proper documents to users as far as possible, so as to satisfy their requests and realize the collaborative relationship.First the article introduces the topic background, the research content and significance, outlines some related concepts about management information system based on B/S modeas well as the function and realize technology of the experimental environment——manufacture enterprise collaborative information network platform. Then makes a description on information retrieval technology, provides the' method to realize the information retrieval sub system based on VSM on the existing collaborative information network platform, and introduces the detailed design mentality: uses vector space model to establish the index sub-vector space of documents and queries separately, carries on the retrieval comparison by computing their similarities. At the same time, relevance feedback and query expansion technology are used together to enhance communication with users and finally to a higher precision and recall are achieved. Afterwards, introduces the involved essential technology in the realization of retrieval sub system, including the patterns and methods to gather and train the collaborative data. Finally simply introducs the realization and the use of the retrieve sub system as well as the test results to the system. From the results of the exercises we can conclude that the combined usage of statistical method and semantic method to realize information retrieval system is full of efficiency and significance in the future.