The semantic web is a global initiative which employs ontologies to offer rich, semantic-based knowledge representation. Concepts in these ontologies are explored to find (dis)similarities between them using (dis)similarity measures. Despite the existence of numerous (dis)similarity measures, none have dynamically determined the quantum of information required to discover (dis)similarities between concepts. In this article, a new, efficient, feature-based semantic dissimilarity measure is proposed where the prime novelty lies in the dynamic selection of the semantic neighourhood (features) of the concepts. The neighbourhood is dynamically selected in accordance with the local density of the concept and the density of the ontology determined by the proposed density coefficient. Further, the proposed measure also scales down the dissimilarity value in accordance with the depth of the concept pair, using the novel Depth Coefficient.

Sathiya Balasubramanian

None Geetha T. V.

International Journal of Intelligent Information Technologies

A New Dynamic Neighbourhood-Based Semantic Dissimilarity Measure for Ontology

A new, efficient, feature-based semantic dissimilarity measure is proposed where the prime novelty lies in the dynamic selection of the semantic neighourhood (features) of the concepts.

A New Dynamic Neighbourhood-Based Semantic Dissimilarity Measure for Ontology

ivySCI AI Smartly Parses PDF, Answers Researchers' Questions, and Helps You Understand Papers in Seconds

Journal Info

Category	Quartile
COMPUTER SCIENCE, INFORMATION SYSTEMS	4