testing framework for OWL ontologies - owl

Question

Is there a testing framework or tool or Protege's plugin for functional testing of OWL ontologies? e.g. to check presence or absence of some axioms in an onology,
to apply some test facts to the ontology and analyse new axioms obtained as a result.

Answer 1

If I understand you correctly you, given an ontology you want to figure out
whether the stated axioms are correct and
whether there are any missing axioms.
(1) OntoComP is a plugin for Protege 4.x. Given an ontology it will, by asking the user questions, either add counter examples or add additional axioms to the ontology. You can read more on this here.
(2) Scenario testing for OWL ontologies defines a method via which you can setup "scenarios" to test that your ontology adheres to functional requirements. This is an approach I have defined in my MSc dissertation and which we have used successfully to validate complex functional requirements for a hotel management group.
(3) You can use the OWL API to test whether the required axioms hold:
OWLOntologyManager manager = OWLManager.createOWLOntologyManager();
OWLDataFactory dataFactory = manager.getOWLDataFactory();
OWLOntology ontology = manager.loadOntologyFromOntologyDocument(ontologyFileIRI);
OWLReasonerFactory reasonerFactory = new ReasonerFactory();
OWLReasoner reasoner = reasonerFactory.createReasoner(ontology);
OWLClass classA = dataFactory.getOWLClass(ontologyIRI + "#A");
OWLClass classD = dataFactory.getOWLClass(ontologyIRI + "#D");
OWLAxiom dSubclassOfA = dataFactory.getOWLSubClassOfAxiom(classD, classA);
reasoner.isConsistent();
System.out.println("D subclass of A = " + reasoner.isEntailed(dSubclassOfA));
The complete example code for this you can find here.

Answer 2

So, this is a challenge to apply testing procedures (functional tests) to a core (i.e. OWL2 driven knowledge base) of an ontology based application.
You can start with mentioned above:
-RDFUnit (https://github.com/AKSW/RDFUnit), but it seems to be very SPARQL specific.
Also things exist like:
-evOWLuator (https://github.com/sisinflab-swot/evowluator).
It is Python based, current version is 0.1.1, and they declare the support for ontology classification, consistency and 'matchmaking' reasoning tasks.
-Scone (https://bitbucket.org/malefort/scone/src/default/). It is quite perspective because allows usage of a controlled natural language, however is still in an early development stage.
And some research efforts are too.
-Again you are free to dive in DIY with OWL API.