Font Size: a A A

Knowledge Fusion & Synchronization Methods Based On Semantic Web Techonlogies

Posted on:2007-07-18Degree:DoctorType:Dissertation
Country:ChinaCandidate:N F XieFull Text:PDF
GTID:1118360185995688Subject:Computer software and theory
Abstract/Summary:PDF Full Text Request
With the rapid development and popularization of the Web, the demand for dynamic data integration, transparent knowledge fusion and information change trace are becoming urgent as more and more information sources appear on the Web. In these cases, the traditional search engines and information integration technologies are insufficient. The reseach on deeper information integration and knowledge synchronization is a promising choice. The advent and development of the Semantic Web brings good opportunities to them. Under such circumstances, this paper makes further analysis and research on an intensional and extensional heterogeneity between distributed information sources, and a knowledge synchronization mechanism. The main contributions of this dissertation are as follows:(1) Providing an XML information integration model based on conceptual spaces. XML is designed to describe data and to focus on what data is, but is poor in complex semantic modeling. Because of its semantic limitation and the rich semantic description capacity of conceptual spaces, the paper proposes an XML information integration based on conceptual spaces, which can solve the semantic integration for XML information sources with XML technologies.(2) Designing multiple strategies to solve the mapping between ontologies. In the Semantic Web, the information can be seen as an instance set of concepts in multiple conceptual spaces. These conceptual spaces were generally built by different information providers independently, so semantic conflicts are unavoidable .Based on an analysis of the exiting ontology mapping methods, the paper adopts linguistics-based and learning-based technologies to semi-automatically build the semantic mappings from local conceptual spaces to a global conceptual space. In the mapping process, the paper makes use of conceptual space information and the instance information coming from the XML information sources.(3) Providing a knowledge fusion model for the Semantic Web. This paper makes further analysis of the deficiency about information process in information integration, so the mixed method based on knowledge fusion and data quality is proposed to make deeper process about data level conflict between different information descriptions about the same object from different sources, which is generally omitted in the current...
Keywords/Search Tags:Knowledge Fusion, Information Integration, Information Synchronization, Concept Space, Ontology mapping, Fusion Rule, Semantic Web, Inconsistencies
PDF Full Text Request
Related items