We define ontology as a formal specification of a vocabulary of concepts, including axioms relating these concepts. A subontology S of ontology N is an extension of the vocabulary of N introducing new concepts, new axioms, and new facts. The ontology translation problem is to translate the concepts and relationships of one ontology (the source) into the vocabulary of another (the target). We believe this is feasible only if both are subontologies of a common ancestor. Hence the first step in translation is to find ancestors of the source and target for which a merged ontology exists, that is, a common framework that relates them. Using this common framework, OntoMerge deductively translates from one vocabulary to another.
An important special case is when the source ontology is a dataset, that is, a set of relationships with no new concepts or axioms. However, our system can handle the more general case. For instance, our ontology library includes a merged ontology for OWL/RDF and WSDL, which are complementary abstract descriptions of web services. Given a subontology of OWL/RDF, OntoMerge can automatically generate the corresponding "grounding" of that subontology into WSDL.
Currently, we have the following merged ontologies available for your use:
More: If you have an ontology written in OWL, you can use our OWL ⇔ PDDL translation service to translate your OWL ontology into Web-PDDL.