A Research Of Search Engine Based On The Semantic Web

Information Retrieval(IR) is a procedure that extracts related information anddocuments from the data sets, while the advent of Web change the way of IR--theobjects of IR transform from constructed data into semi-constructed and non-constructed data. With the increasing accumulation of data on the Web and the rising demand of high quality of the retrieval results, it has been very difficult for the traditional Web IR technologies to satisfy the task of highly desirable results. The Semantic Web is an effective approach to solve the problem of high-quality IR, distributed computation based on the Internet and knowledge discovery and reference. To the search engine, the Semantic Web plays an important role in the development of a new generation of search engines, meanwhile, it improves the information retrieval on the Web to a high quality and intelligent level.This thesis mainly discuss how to improve the precision and relativity of the retrieval, the IR model and algorithm, semantic information model, and so on, which combine organically the RDF, RDFSchema, Ontology with the search engine, and put forward with some effective ways to enhance the quality and efficiency whereby the consequence of the retrieval is satisfiable.On the basis of the completion of the theory research on the semantic web we construct a prototype-ISeek, which includes RDF/XML parser, RDF documents store subsystem, reference subsystem, query subsystem, and administrationsubsystem,constituting a prototype of a search engine.