A Selection of a Reduced Set of Similarity Measures for Concept Mapping in the Ontological Enrichment Process
Résumé
The similarity measures are often used in ontological engineering. Potential applications of these measures include information retrieval and clustering, data mining, knowledge discovery and decision-support systems that use ontologies. In our context, we want to enrich an existing core ontology that models satellite image content by new concepts coming from related geographical ontologies. Our main goal is to select a reduced set of similarity measures that captures all the information we have on similarity between two concepts. In this paper, we first present a literature review of existing similarity models and measures, then we carry out an experimental analysis in order to select a reduced and relevant set of similarity measures.