Assessing the quality of ontology

In the Information Sciences an ontology specifies the conceptual structure of a knowledge domain through its vocabulary and its meaning, or semantics. Examples of ontologies include database schemas, taxonomies and library classification schemes. Computer-based ontologies specify a common understanding of a domain both for its different human users and for automated information processes such as interpretation, aggregation, manipulation, etc. As an ontology is fundamental to systems in an organisation and to communication between organisations, its quality is critically important for the sound operation of those systems. But there is a problem in that there seems to be no systematic basis or method for conducting an assessment of the quality of an ontology. There is an abundance of literature proposing ideas about what constitutes a 'good' ontology, but it is diverse, incoherent and untested. This paper reports research which reviewed the literature, extracted some 220 quality criteria, formulated these into an assessment instrument and partially demonstrated the instrument in a case study. The demonstration showed some of the difficulties associated with assessing quality, in particular, the difficulty of grounding the criteria in the actuality of the case and in interpreting the results of an assessment.