| With the rapid development of information technology and Internet, the information resources of the best evidence of EBM have been increasingly rich. However, the informationalization, automatization, intelligentization, and marketization of EBM is not very developed at present, there are significant gaps in meeting social needs. Therefore key technologies for developing the infromationalization of EBM should be in-depth studied. The degree of information, the level of intelligence and the dgree of automation should be promoted, and the feasibility and practicality of generating the evidences of EBM should be improved through utilizing intelligent technologies. In the ultimate, the products and marketization of the outcome of the EBM technologiy need to strive to achieve.The main work of the thesis includes:Firstly, the structure of searching engine, technologies on Internet robot and Lucene are studied. The Google's PageRank algorithm is analyzed and some disadavantages of this method are pointed out. Those disadvantages include preferring to old pages, ignoring special sites and inaccurate judge of hyperlinks from one page. Some improvements of Lucene's sorting algorithm are presented. The experiments show that the consideration on evaluating the importance of pages can make an improvement over the original algorithm.Secondly, Loincn -a full-text retrieval search engine system is designed. This system consists of the four main parts of the searching engine: spider, indexer, searcher and UI. The experiments show that this search engine meets the goal of providing users with EBM data query functions and automatic article summaries, etc.Thirdly, a Web-based EBM informationalized platform is designed. This platform integrates webpage intelligent collection, filtering, classification, sumarization, systematic review and other functions, it facilitates operators and improves the work efficiency, and it makes the output of the project praticable and feasible in social. |
PDF Full Text Request