Research On The Key Technologies Of Heterogeneous Resources Integration Based On Semantic Grid

With the development of computer technology and social progress, social information has sharpply increased in the amount and the information continues to short time. It is relatively limited capacity to receive information for the people. How to organize network resources to provide search services transparent to the user becomes an urgent problem.Grid technology has been used to integrate the distributed resources to provide integrated information services to users as a variety of high-speed Internet facilitate, but semantic heterogeneity can not be resolved due to the grid resources lack of semantic, it is difficult to achieve efficient resource sharing. Semantic grid technology which integrated by the advantages of semantic Web technology and grid technology has become a hot integration of heterogeneous resources.Taking the digital library resources as the research background, this thesis use semantic grid to research integration issues for heterogeneous resources, and submit an integrated architecture of heterogeneous resources which based on the Semantic Grid. This architecture uses multi-layer framework for heterogeneous integration of resources, and it takes OGSA-DAI middleware as logic integration layer, also it established ontology mapping layer, Semantic Grid service layer to deal with intermediate links such as Grid service agent and semantic translation services.This thesis makes DC metadata standard as the criterion for standard literature resource database. Based on this standard, two level ontology mapping models was established.With these models the process of semantic service is analyzed, such as semantic indexing, semantic registration, semantic reasoning and semantic query. Using OWL-S ontology language standardization the user's needs can improve the efficiency of semantic retrieval.Finally, it has implemented a cross-library retrieval prototype system for digital library platform supported by the literature resources. By using the Semantic concept similarity calculation algorithm to extend key words, experimental results show that the proposed integrated model can improve the efficiency of document retrieval and recall.