Research On Information Integration For Teaching Management Based On Ontology

Along with the development of network and requirement of teaching management, more and more heterogeneous datasources had been produced, the need of solving the integration problem of these heterogeneous datasources was becoming more important. This paper mainly imported the concept of ontology to realize the integration of heterogeneous data. Utilization of ontology did well in solving the problem of semantic heterogeneity of information integration, however the other technologies of data integration could hardly achieve it.The paper dicussed the current technologies of data integration, then introduced the concept of ontology, and indicated the advantage of applying ontology to the integration of heterogeneous data,following the description of several ways of using ontology in information integration, then emphasized on presenting the framework of ontology based teaching management information integration, built a global ontology for all datasources and a local ontology for every datasource, at the same time set up the mapping between ontologies and datasources.Relational databases are the main information sources of the OTMIIS , and the extraction of schema information from relational databases is one of the basic techniques for transferring the current Data into the OWL ontolgogy. In this paper , a prototype tool Extractor , as well as its design ideas and implementation techniques based on JDBC API , is introduced , and the schema information of relational databases is analyzed. Case studies show that the JDBC API based schema extraction method is feasible, and that Extractor is effective.A method for semantics preserving translation from ER schema to OWL DL ontology is presented. Based on the formalization of ER schema , the method establishes precise conceptual correspon dences between an ER schema and an OWL DL ontology and realizes the schema translation via a translation algorithm which follows a set of predefined mapping rules. Theoretic analysis shows that the method is semantics-preserving and effective. Algorithmic implementation and case study demonstrates farther that the fully automated machine translation is realizable. The proposed method is original and will doubtless facilitate the development of Web ontologies and the realization of semantic interoperations between existing databaseapplications and the Semantic Web.