Analysis of Ontology Quality Dimensions, Criteria and Metrics

Abstract

Ontology quality assessment needs to be performed across the ontology development life cycle to ensure that the ontology being modeled meets the intended purpose. To this end, a set of quality criteria and metrics provides a basis to assess the quality with respect to the quality requirements. However, the existing criteria and metrics defined in the literature so far are messy and vague. Thus, it is difficult to determine what set of criteria and measures would be applicable to assess the quality of an ontology for the intended purpose. Moreover, there are no well-accepted methodologies for ontology quality assessment as the way it is in the software engineering discipline. Therefore, a comprehensive review was performed to identify the existing contribution on ontology quality criteria and metrics. As a result, it was identified that the existing criteria can be classified under five dimensions namely syntactic, structural, semantic, pragmatic, and social. Moreover, a matrix with ontology levels, approaches, and criteria/metrics was presented to guide the researchers when they perform a quality assessment.