Research And Implement On A Semanteme-Oriented Integration Framework For Heterogeneous Data

With the development of the computer technologies and the popularization of the computer network, especially the development and maturation of XML and related technologies, the integration of the heterogeneous data has become one of the heat points in the process of informationization of the modern companies. Based on the related technologies of Semantic-Web, this paper proposes a heterogeneous data integration method based on semantics.Based on the current data integration technologies, this paper proposes XML and RDF/RDFS based heterogeneous data integration approach which imports the"middleware-wrapper"systematic structure. This approach uses XML Schema to represent the data model of various autonomy data sources and through RDF/RDFS, this paper builds the semantic model based on the data model. In this way, the semantic heterogeneous problem between data resources has been well solved. At the same time, this paper makes research on the approaches of building global semantic model and extracting the local semantic model from RDS. And this paper also proposes the mapping approaches from local semantic model to global semantic model and he mapping approaches from relation data sources to local semantic model.RDQL(RDF Query Language)is used as the query language of the data integration system. RDQL is a tree query language. By parsing the query request, a query tree can be constructed. Then by making use of the mapping rules from local semantic model to global semantic modle, the query tree is decomposed into sub-trees, and the query decomposition algorithm is also presented. The sub-query is then rewritten to query language related to data sources.In the system implementation, this paper uses Java programming language, Jena, Saxon and other open sources software to build a general heterogeneous data integration frame model. Also, this paper tests the proposed approaches with a classical test case.