| With the development of the information technology and its relevant industries, web database has become a large information resource depository. These informaton can not be searched by the traditional search engine. Users have to fill out forms and submit queries to obtain the data. We call these information as Deep Web. In order to help users to find out what they want, we need to integrate the Deep Web information and provide information retrieval services.Schema matching technology is needed in the whole process of Deep Web information integration. This paper firstly focuses on technology of Deep Web and schema matching, and then proposes an ontology-based schema matching method to solve problems of interface integration and data annotation. Finally Deep Web object-level query interfaces integration archetypal system is presented. The main work is summarized as follows:(1)Firstly, introduce relevant knowledge of Deep Web and Ontology, then analysing and comparing of the traditional schema matching and Deep Web schema matching.(2)Propose an ontology-based schema matching method for the Deep Web interfaces integration. The method uses ontoloty to help improve the precision of schema matching results, based on the three methods of labeling, value domain and name of control, uses the similarity combination strategy and the highest similarity priority algorithm to solve the schema matching problem in interface intergration.(3)Propose an onology-based schema matching method for the annotation of the Deep Web. The method uses ontology as the whole schema, solves the problem of semantic heterogeneity within interfaces. The paper improves precision by cross validating matching results between ontology, interface module and result module.(4)Design a Deep Web object-level query interfaces integration system according to above technology.Finally we design experiments to implement mentioned algorithms and technology, and experiment results show these methods are effective. |
PDF Full Text Request