Enhanced Search Method for Ontology Classification

keywords: Ontolgy, DL reasoning, tableaux algorithm, classification, subsumption test, top search, bottom search
The web ontology language (OWL) has become a W3C recommendation to publish and share ontologies on the semantic web. In order to infer implicit information (classification, satisfiability and realization) of OWL ontology, a number of OWL reasoners have been introduced. Ontology classification is to compute a partial ordering or hierarchy of named concepts in the ontology using the subsumption testing. Most of the reasoners use both top-down and bottom-up searches using subsumption testing for ontology classification. As subsumption testing is costly, it is important to ensure that the classification process uses the smallest number of tests. In this paper, we propose an enhanced method of optimizing the ontology classification process of ontology reasoning. Our work focuses on two key aspects: The first and foremost, we describe classical methods for ontology classification. Next, we present description of the enhanced method of optimizing the ontology classification and the detailed algorithm. We evaluate the effect of the enhanced method on four different types of test ontology. The enhanced search method shows 30 % performance improvement as compared with the classical method according to the result of the experiment.
reference: Vol. 28, 2009, No. 6, pp. 795–809