A Transformation Approach for Classifying ALCHI(D) Ontologies with a Consequence-based ALCH Reasoner.
Weihong Song, Bruce Spencer, Weichang Du · 2013
Consequence-based techniques have been developed to provide efficient classification for less expressive languages. Ontology transformation techniques are often used to approximate axioms in a more expressive language by axioms in a less expressive language. In this paper, we present an approach to use a fast consequence-based ALCH reasoner to classify anALCHI(D) ontology with a subset of OWL 2 datatypes and facets. We transform datatype and inverse role axioms intoALCH axioms. The transformed ontology preserves sound and complete classification w.r.t the original ontology. The proposed approach has been implemented in the prototype WSClassifier which exhibits the high performance of consequence reasoning. The experiments show that for classifying large and highly cyclic ALCHI(D) ontologies, WSClassifier’s performance is significantly faster than tableau-based reasoners.