Abstract: Aiming at the problem of low mapping accuracy and efficiency caused by semantic heterogeneity in ontology mapping, especially in large-scale heterogeneous ontology mapping, we  proposed an  ontology mapping method based on node semantic similarity (NSS). Firstly, we studied  key technologies such as web-based ontology parsing and representation, automatic ontology partitioning, rapid recognition of similar sub ontologies, and node semantic based sub ontology mapping. Secondly, the experiments were conducted on the conference ontology set in the ontology alignment evaluation initiative (OAEI) evaluation datasets. The results show that the proposed method outperforms traditional mapping methods in performance and has higher accuracy than fragment based mapping methods.

Key words: semantic similarity, ontology mapping, ontology partition, ontology alignment evaluation initiative, precision, efficiency