Font Size: a A A

Data Semantic Retrieval System Based On Ontology Research And Design

Posted on:2007-07-17Degree:MasterType:Thesis
Country:ChinaCandidate:Q HuFull Text:PDF
GTID:2208360185973489Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
With the development of the technology of network and software, especially with the development of the Internet and Intranet, the most of information provided to users are mostly not only structured, but also unstructured. Kinds of information appear rapidly, the complexion of complicated data resources which are in different format, different date, and different application is formed step by step. It is an undoubted valuable question for discus that How to set up an expression mechanism of kinds of information and build a uniform plat to dealing with and retrieving information, so practical application of information could be satisfied.The traditional retrieval based on the keywords matching has the limitation of efficiency. Users are not satisfied because of the error and miss.A new retrieval method is provided by Semantic Web and Ontology, through which the limitation of the traditional retrieval system can be resolved. At the present time information retrieval based on Ontology is focus of the research and the application of Ontology.In this paper, firstly, the theory and the advantages of applying are introduced. Secondly, two questions about retrieval based on Ontology are needed to be resolved, they are how to create XML documents automatically which is connected with the retrieval system and how to improve the retrieval efficiency in using Ontology technology.In part of system design the total design of the distributed retrieval system is provided, and then an algorithm of abstracting file information is also presented. That algorithm is based on method of maximal matching. It takes the Ontology of relative fields that enterprises and the organizations concerned with as words segment dictionary, searching literally, longest matching, deleting the non-use word, and then recognising Ontology by marking the key word to segment words. Finally, appearance frequencies (times/text length) of ontology words are calculated and stored in XML document for search.As to the retrieval algorithm design based on Ontology, firstly, a similarity (word, file) function about calculating the similarity degree of the key words and the files is...
Keywords/Search Tags:Ontology, retrieval, XML
PDF Full Text Request
Related items