The OWL 2 Web Ontology Language is the principal standard for building formal ontologies on the Semantic Web. As a W3C Recommendation, it provides a rich set of constructs for defining classes, properties, individuals, and their relationships with formally defined semantics. OWL 2 is one of the core building blocks of the Semantic Web technology stack, enabling applications to perform automated reasoning over knowledge represented as ontologies.
Background
The original Web Ontology Language (OWL 1) was published as a W3C Recommendation in 2004 by the Web Ontology Working Group. It was designed to facilitate ontology development and sharing via the Web, with the goal of making Web content more accessible to machines. OWL 2 is an extension and revision developed by a follow-on W3C OWL Working Group, published as a W3C Recommendation on 27 October 2009 with a Second Edition on 11 December 2012. OWL 2 maintains nearly complete backwards compatibility with OWL 1 -- all OWL 1 ontologies remain valid OWL 2 ontologies with identical inferences in practical cases.
Purpose & Scope
OWL 2 ontologies provide classes, properties, individuals, and data values, and are stored as Semantic Web documents. They can be used alongside information written in RDF, and are primarily exchanged as RDF documents. The language supports diverse use cases from biomedical knowledge representation to enterprise data integration and Linked Data applications. An OWL 2 ontology can be thought of either as an abstract structure or as an RDF graph, with the Mapping to RDF Graphs specification defining the relationship between these two views.
Key Features
OWL 2 adds several capabilities beyond OWL 1:
- Keys for identifying individuals
- Property chains for composing relationships
- Richer datatypes and data ranges
- Qualified cardinality restrictions
- Asymmetric, reflexive, and disjoint properties
- Enhanced annotation capabilities
- Disjoint union of classes
Syntaxes
OWL 2 supports multiple concrete syntaxes for serialization and exchange:
| Syntax | Status | Purpose |
|---|---|---|
| RDF/XML | Mandatory | Primary interchange format; must be supported by all conformant tools |
| OWL/XML | Optional | Easier processing with XML tools |
| Functional Syntax | Optional | Exposes the formal structure of ontologies |
| Manchester Syntax | Optional | Human-readable syntax used in ontology editors |
| Turtle | Optional | Readable RDF triple serialization |
Semantics & Profiles
OWL 2 offers two alternative semantic interpretations: the Direct Semantics (compatible with the SROIQ description logic, a fragment of first-order logic with useful computational properties) and the RDF-Based Semantics (fully compatible with and extending RDF Semantics). A correspondence theorem ensures that inferences drawn under the Direct Semantics remain valid under the RDF-Based Semantics.
Three computational profiles provide trade-offs between expressiveness and performance:
- OWL 2 EL -- Polynomial time algorithms for all standard reasoning tasks; suitable for very large ontologies where expressive power can be traded for performance guarantees
- OWL 2 QL -- Conjunctive queries answerable in LogSpace (AC0) using standard relational database technology; suitable for lightweight ontologies organizing large numbers of individuals
- OWL 2 RL -- Polynomial time reasoning using rule-extended database technologies operating directly on RDF triples; suitable for operating directly on data in RDF form
Specification Suite
The OWL 2 specification comprises thirteen documents:
| Part | Type | Document |
|---|---|---|
| 1 | User Guide | Document Overview |
| 2 | Core | Structural Specification and Functional-Style Syntax |
| 3 | Core | Mapping to RDF Graphs |
| 4 | Core | Direct Semantics |
| 5 | Core | RDF-Based Semantics |
| 6 | Core | Conformance |
| 7 | Spec | Profiles (EL, QL, RL) |
| 8 | User Guide | Primer |
| 9 | User Guide | New Features and Rationale |
| 10 | User Guide | Quick Reference Guide |
| 11 | Spec | XML Serialization |
| 12 | Spec | Manchester Syntax |
| 13 | Spec | Data Range Extension: Linear Equations |
Governance & Maintenance
OWL 2 is maintained by the World Wide Web Consortium. The current Second Edition was published 11 December 2012 under the W3C Patent Policy. The OWL Working Group included members from Oxford University, Bell Labs Research, FZI, University of Manchester, University of Aberdeen, RPI, and many other institutions.
Notable Implementations
OWL 2 is implemented across a broad ecosystem of tools and applications. Major reasoners include Pellet, HermiT, FaCT++, and ELK. The Protege ontology editor from Stanford is the most widely used tool for building OWL ontologies. OWL 2 ontologies are deployed extensively in biomedicine (the Gene Ontology, SNOMED CT), enterprise knowledge graphs, and Linked Data applications.
Related Standards
- RDF -- The foundational data model that OWL 2 extends
- RDFS -- RDF Schema, a simpler vocabulary language that OWL 2 builds upon
- SKOS -- A lightweight vocabulary for knowledge organization systems, complementary to OWL
- SPARQL -- The query language for RDF data, used to query OWL 2 ontologies