The Design, Extraction And Application Of Ontology Metadata

Ontology is widely referenced to computer field, especially to the field of artificial intelligence to solve problems relating to knowledge representation and knowledge organization. As the building and application of ontology is studying more and more deeply, the number of ontology swells. There are many differences among different ontologies in such aspects as scale, the integrity and normalization. Falcons grabs 312480 ontologies from global network, providing query service and documentation. Swoogle developed by the University of Maryland does the work similarly. But statistical analysis of these differences can’t be found currently which makes it necessary for us to discuss the standard of ontology description.At present, ontology construction, evaluation and application are lack of norms and standards. This article analyzes the structure of ontology, introduces the Dublin core element set and designs ontology metadata framework. For each type of metadata, this paper puts forward the extraction process and algorithm. On the basis of metadata, ontology of Falcons system is analyzed. And finally, OWL ontology classification experiment is carried out, laying foundation for the further study of the ontology.The contribution of this paper lies in putting forward a metadata framework not staying in the theoretical level but exploring the possibility of extracting metadata, analyzing the relationship between ontology model and description language when using Jena and summarizing reasons for unable to extract metadata. Based on associated resources item of the ontology metadata framework, this article analyses the entity set and XML namespaces, trying to classify the OWL ontologies from Falcons query system by semi-automatic way referring to the assertion set, which provides a more practical idea for ontology classification.