Medical Information Classification And Retrieval System

In the Internet age, various types of information services are increasingly rich, data overload problem is becoming increasingly prominent. Medical information is closely related to human health, and a large amount of medical information has emerged on the internet. The general search engine results of medical information retrieval and intelligent recommendation mixed together, the urgent need for medical consultation unified precise classification of technical support. At present,the main carrier of medical information is text, so the text automatic classification based classification and retrieval technology in medical information retrieval, it has important practical value for medical information fast acquisition and utilization.In this thesis, the classification and retrieval of medical information is studied. Based on the text classification algorithm and the open source search engine framework, the medical information collection, automatic classification and retrieval system is realized. The main work is as follows:(1) According to the current mainstream medical information page structure, using the web crawler technology, the medical information collection is realized.(2) For multi-label classification of medical information of medical information research on the key technology of text classification, analysis,comparison and selection of MMSeg4j test Chinese segmentation, chi square test and SVM classification algorithm, using two times classification strategy to achieve classification.(3) Based on the text retrieval algorithm, a fast retrieval technology for medical information data sets is studied.(4) Based on the LIBSVM and Solr open source search framework,develop a medical information automatic classification and retrieval system, can realize the automatic classification of medical information, and provide fast retrieval and classification of interactive web pages to achieve information display.