Ontology Quality by Detection of Conflicts in Metadata.
Ismailcem Budak Arpinar, Karthikeyan Giriloganathan, Boanerges Aleman-Meza · 2006
ontologies continue to be an important component in techniques and applications in semantic technologies. Thus, it is necessary to evaluate their quality. Our focus is the detection of conflicting information (within an ontology) as a criterion to improve the quality of an ontology. We describe different types of conflicts and propose a rule-based approach by which human experts can define conditions that signal a conflict in data. These rules (represented using RuleML) are used to automatically detect conflicts in populated ontologies. We describe a prototype application and evaluate the applicability of this approach.