Estimation of Semantic Similarity between Concepts using Multiple Ontologies (Wordnet and Mesh) FOR Biomedical DATA

The majority of the intelligent knowledge-based applications include elements for the purpose of measuring semantic similarity stuck between terms. A lot of the available semantic similarity measures that make use of ontology structure as their chief source cannot determine semantic similarity among terms and concepts by means of multiple ontologies. Hence, this research looks for a new approach to determine the semantic similarity among the biomedical concepts by means of multiple ontologies. In this work, a novel resemblance assess is proposed and it is combining both super concept of the assessed ideas and their general certainity feature. This feature takes the deepness of the Least Common Subsumer (LCS) of two ideas and the deepness of the ontology by the way to attain further semantic evidence. The similarity in the midst of two concepts is a weighted sum total of the resemblances of the two features among them. The features are Data content similarity and Presentation style similarity. The weight is calculated by using rank aggregation criteria. The similarity is measured based on some rules and assumptions. The similarity measure is based on Information Content (IC) and context vector. Subsequently, the proposed measure was assessed relative to human specialists’ ratings, and evaluated against the existing schemes of biomedical terms by means of two standard ontologies (WordNet and MeSH). WordNet was used as primary general ontology and MeSH was used as secondary ontology. The investigational outcomes proved the effectiveness of the proposed scheme, and demonstrated that proposed similarity measure provides the most excellent complete outcomes of correlation with individual ratings.