Semantic Similarity Measures

Semantic web, the vision of Tim Berner's Lee, is an effort to enhance current web, by giving ¿well-defined meaning¿ to the content of Web pages , better enabling computers and people to work in cooperation. Ontology is at the heart of Semantic Web describing the concepts, their relationships and properties within their domain. ¿Reusability Approach to Ontology Building¿ utilizes the online ontologies and proves to be better when complete domain knowledge is not available at the beginning. To support interoperability, Mapping and Merging between ontologies, finding the Semantic Similarity between concepts of ontology play a key role. It is essential to handle the problem raised during semantic similarity computation, when the concepts for which the similarity is being found are subsumed by multiple super concepts in other words represent multiple inheritance. Two new measures ontology based and Information Content based have been applied on concept pairs of benchmark data sets to evaluate their efficiency in similarity calculation. The experimental results proved the effectiveness and superiority of the two similarity measures compared with the existing similar techniques.