Research And Implementation Of XML Semantic Information Extraction Based On Domain Ontology

XML(Extensible Markup Language) as W3C standard format language of data description and data exchange has been widely used.Either Semantic Web or Web Services uses XML as its Standard Format for data presentation and data exchange.XML has develop into the main media for data presentation and data exchange in the information application or other fields.However,XML could only expresses data syntax,not expresses formal semantics,semantic information that is implicit in XML document is of significance to people,but is hard to understand for the computer.To achieve the goal of computer's understanding of documents information and automatic processing,documents data should includes definite semantic level information.This thesis put forward a methods achieveing the xml semantic information extraction over domain ontology,which contains two parts.That is the construction of domain ontology and the xml semantic information extraction over domain ontology.In the first part,this thesis comparative studies of the building tools for domain ontology and building methods,and proposed a method of prototypes iteration for construction of domain ontology that introduces the iteration model of software life cycle in software engineering into the ontology construction process.This thesis builds a domain ontology for tariff and checking of water transportation.In order to reduce the workload of ontology construction,this thesis uses the prot(?)g(?) ontology development tools of Stanford University and Racer inference engine of Hamburg University as a ontology validation tool.In the second part,this paper presented an xml semantic information extraction model over domain ontology,which includes the ontology analysis module,xml data source importing module,semantic tagging module and semantic extraction module, and gives a detailed analysis of the realization of the four modules.Through analysis of the XML tree Model and RDF graph model in-depth,this thesis proposes a extraction algorithm converting XML into RDF and shows verification results via prototype system of XTR Service.The main technology used includes Jdom API, Jena API,as well as the depth-first traversal algorithm of general tree.The research work of this paper has basically worked out the problem of semantic information extraction of XML,especially the implied semantic of XML, with the help of actual research projects and the guidance of domain ontology.