Research On Semantic Heterogeneity Elimination For Information Integration

Sharp development of information technology and extensive application of internet produces massive distributed heterogeneous information sources, which has different forms, representations and storage format. It is hard for consumers to use these heterogeneous information. Efficient usage of these sources calls for integration of distributed, heterogeneous and autonomous information sources. Much attention has been focused on the systems, structure and grammars heterogeneous, seldom on the semantic heterogeneously, so semantic heterogeneity remains to be solved, while the problem of heterogeneous systems, structures and grammars are handled by traditional information integration techniques, eliminating semantic heterogeneity turns out to be important to handle the increase of information sources and to enforce techniques.In this paper, ontology technique is employed in heterogeneous information integration, and a heterogeneous information framework is given Main contributions of the paper are as follows: First, the way to map relational database and XML documents to ontology is proposed and evaluated. Moreover, calculation methods of comprehensive similarity based upon data levels and tree structure levels are proposed, based on which semantic conflicts elimination is executed and a global ontology is constructed, resulting in an unified view.As is analyzed in thesis, the proposed method solves semantic heterogeneity problem effectively.