Identifying Relevant Concepts Attributes to Support Mapping Maintenance under Ontology Evolution
Abstract
The success of distributed and semantic-enabled systems relies on the use of up-to-date ontologies and mappings. However, the size, quantity and dynamics of existing ontologies demand huge maintenance effort pushing towards the development of automatic tools for supporting this laborious task. This article proposes a novel method, investigating different types of similarity measures, to identify concepts' attributes that served to define existing mappings. The obtained experimental results reveal that our proposed method allows to identify the relevant attributes for supporting mapping maintenance, since we found correlations between ontology changes afecting the identified attributes and mappings change.