Research Of Semantic Grid Based Information Retrieval Techniques

The fast development of information technology has provided us a wide platform for resource share. One can use information retrieval tools to find the relevant information, this, to some extent, solves the problem of information classification and retrieval. But, this conventional key words-based retrieval pattern neglects semantic meanings contained in the key words, and has the low pertinency factor and recall factor, is poorly catering for the application requirements. With the help of semantic grid and ontology technology, a new intelligent method of information retrieval has developed. This thesis focuses on the information retrieval technology base on semantic grid and contributes to:(1) Posed system design architecture of computer literature information retrieval based on semantic grid, introduced key modules contriving including interrelated ontology, data source choosing, information retrieval agent, semantic intersection, reasoning engine and service wrap;(2) Put an emphasis on the study of ontology-driven literature retrieval related technology on semantic grid environment: how to build correlative ontology; how to indeed understand user's search requirement based on ontology and logical reasoning and how to put the retrieval requirement to appropriate data sources to implement this task.(3) Implemented a prototype system SeirGRID on the grid platform of Globus toolkit 3.0. Making use of ontology edit tool protégé, ontology was built including service ontology, computer literature ontology, information source ontology and retrieval constructing ontology. Concerning data sources , IEEE,SCI,Elsevier Science and CNKI were arranged for candidates. On the condition, key words-based and concept-based information retrieval methods were both accomplished, and the latter implemented semantic graphic and fuzzy retrieval navigation, realized logical reasoning and semantic extension, fulfilled grid service wrap for retrieval task, and developed information retrieval agent to find fitful data sources, and accomplished retrieval service employing data source ontology.(4) Made a conclusion, pointed out the disadvantages of the system, and expected semantic grid based ontology will take on an exciting look in the field of information retrieval.