An Ontology Concept Update Method Based on Hybrid Semantic Similarity

With the change of information, the existing ontology cannot satisfy users' requirements. In order to realize the extension of ontology concepts and reduce the excessive dependence on domain experts, an ontology concept update method is proposed. Based on WordNet, the calculation of IC(Information Content) is improved, and then we propose a new hybrid measurement to calculate semantic similarity. Due to the data and object properties of ontology are ignored by the semantic similarity methods, property similarity is added to make an adjustment. Through experiments and comparison, the Pearson correlation coefficient between the improved semantic similarity method and the standard datasets is higher than other related methods, and the result is closer to human subjective judgment. The ontology concept update method is used to analyze a constructed ontology of academic conference. The results show that this method can update the ontology concept, and has certa.