A Concept Similarity Computation Based On Multi-Property Ontology

As we know, the core purpose for the use of ontologies is the exchange of data not only at a common syntactic, but also at a shared semantic level. On the WWW more and more ontologies are constructed and used, beginning to replace the old-fashioned ways of exchanging business data via standardized comma-separated formats by standards that adhere to semantic specifications given through ontologies. With this upswing and beginning widespread usage of ontologies, however, new problems are incurred. Ontology engineers or users frequently have a core ontology that they use, e.g., for browsing or querying data, but they need to extend it with, adapt it to, or compare it with the large set of other ontologies. And research areas such as semantic search, information integration and ontology mapping all need to recognize and estimate the semantic between different ontologies in order to share knowledge. For the task of detecting and retrieving relevant ontologies, one needs means for measuring the similarity between ontologies.No matter during the ontology mapping or the ontology evolution, the attribute-based concepts similarity computation is the key point of the ontology research. For the complex ontology, the concepts which are presented in the ontology have the multi-attribute character. The inheritance and the particularity of the attribute will directly affect the relevance of the concept and decide the precision of the semantic inference of the concept.In this paper, we propose a novel multi properties-based concepts similarity calculation under the theory of evolution and taxonomy. The dialectical relationship of the ancestral characters and new-born characters in evolution will be brought in the transmission of the ontology concept property. Based on the evolution of the ancestral character and new-born character of the concept property, the changes among categories will be analyzed to improve the accuracy of the similarity measure of the multi-properties concept. In order to show concept similarity calculating accuracy, we build tea multi-property ontology. To study the tea ontology properties evolution method, we provide various properties computation priority algorithms. Finally we compare calculating result four kinks of method, and show our method that calculating accuracy is high, and in accordance with tea domain ontology expert's opinion.