Semantic Similarity Measures on Different Ontologies: Survey and a Proposal of Cross Ontology based Similarity Measure

Semantic similarity relates to computing the similarity between conceptually similar but not necessarily lexically similar terms. Semantic-similarity measures quantify Concept Similarities in a given ontology. Typically, it is computed by mapping terms to ontology and by examining their relationships in that ontology. In this paper, a comparative study on different measures such as path based, information content based, feature based and hybrid similarity measures is done for identifying semantically similar concepts in ontology. The focus is on more than one ontology methods since it is interesting than the single ontology and semantic similarities are calculated between terms stemming from different ontologies (WordNet and MeSH in this work). The purpose of this survey is to explore how these similarity computation methods could assist to improve the retrieval effectiveness of Information retrieval models based on Cross Ontology.