Abstract:

Concept similarity evaluation is widely used in concept classification, concept gathering and concept
merging. Previous single-ontology-oriented models were centered on matched concept characteristics and semantic
analysis. Since they did not take enough factors into consideration, the obtained results were inaccurate, therefore
of less value for exploring relations among concepts. This paper makes the matched concept characteristics more
complete, takes characteristics which are not processed among concepts as similarity, assigns different weights according
to the semantic hierarchy of the concepts, and proposes an improved concept similarity calculation model
with high precision. Experimental results show that the improved model is more accurate.

Key words: concept, characteristic, semantic, similarity, ontology