Research On Large-Scale Ontology Partition And Mapping

The goal of Ontology Mapping is to construct semantic bridge between different ontologies. Because large-scale ontologies have large number of concepts and complicated relations, it is very difficult to implement mapping between them. The thesis focuses on research on this issue.Firstly, the purposes of researching and developing SNAX, backdrop of Semantic Web research domain and current research situation of Ontology Mapping have been introduced.Secondly, according to the structure and features of large-scale ontology, a method of clustering and partition of class hierarchies based on CURD algorithm has been put forward by applying Vector Space Model(VSM). This method constructs vector space by words stemmed from concepts' information, and puts concepts in this space by using Semantic Diffuse algorithm, then carries out concepts cluster by CURD algorithm. The matching between blocks can be carried out by computing the similarity of block's central points.Finally, based on the research above, Large-scale Ontology Partition and Mapping(LOMAP) system has been designed and implemented. Moreover, the mapping effective of the system has been tested through using benchmarks, ontology set offered by OAEI. The precision, recall, and F-testing values have been statistically analyzed. The results show our approaches are comparatively good.