Mapping SBVR to OWL2

This paper proposes a mapping of SBVR¹ vocabularies to a combination of OWL2² elements and annotations. The goal of this work is to define a reversible mapping: an SBVR vocabulary can be mapped to OWL2 and back again without loss of semantic information. That is, given a transformation chain V1 --> O1 --> V2, where V1 and V2 are SBVR vocabularies, and O1 is an OWL2 ontology, the semantic meaning of V1 and V2 are equivalent. This does not mean that V1 and V2 are textually identical because their format (e.g. Structured English versus graphical), syntactic structure, and lexical details (e.g. white space) may differ. It does imply that in a transformation chain V1 --> O1 --> V2 --> O2, the two SBVR vocabularies have the same meaning, and the two OWL2 ontologies have the same meaning. This also means that the OWL2-based transformation offers a viable format for exchanging SBVR vocabularies between two SBVR tools, as an alternative to the XMI-based format specified in SBVR clause 13.

The scope of this work is SBVR vocabularies, excluding behavioral rules.

The mapping described here depends upon two machine readable files, both of which are available as open source:

• purl.org/sbvr2owl/SBVRTaggedText.xsd – the XML schema that is described in section 4.1 of this document. Referenced in this document by the prefix sbvr-tt.
• purl.org/sbvr2owl/VocabularyDescription.owl – an OWL ontology that implements the OWL annotations and entities that are described in sections 4.2 and 4.3 of this document. Referenced in this document by the prefix sbvr-vd.

This mapping has been applied to the Date-Time Vocabulary (DTV) to generate a set of DTV ontologies in OWL2.

¹SBVR is the Semantics of Business Vocabulary and Rules specification from the Object Management Group (OMG), available at http://www.omg.org/spec/SBVR.
²The OWL2 Web Ontology Language standard of the World Wide Web Consortium (W3C), available at http://www.w3.org/TR/owl2-overview/.