Semantic Similarity Measurement Model Based on Geographic Ontology

Abstract

Traditional keyword based information retrieval methods can not understand the semantic of information. Ontology is the foundation of the Semantic Web. However, the geographic ontology construction studies are not thorough enough. Similarity measurement in geographic information search, particularly in query expansion, plays an important role, while the existing semantic similarity measurement models only focus on single impact factor, which leads to poor convergence performance and lower calculation accuracy, and they are greatly influenced by subjective consciousness. In this paper, the building of geographic ontology is researched. According to four impact factors, a semantic similarity measurement model based on geographic ontology is proposed. The experiment results indicate that this model has better performance than classical models, meaning that the semantic similarity of proposed model is more consistent with the expert experience in field.