FOUNDATIONS OF SEMANTIC WEB TECHNOLOGIES Ontology Editing - PowerPoint PPT Presentation

FOUNDATIONS OF SEMANTIC WEB TECHNOLOGIES Ontology Editing Sebastian Rudolph Dresden, July 2

Content Overview & XML 9 APR DS2 Hypertableau II 7 JUN DS5 Introduction into RDF 9 APR DS3 Tutorial 5 11 JUN DS2 RDFS – Syntax & Intuition 12 APR DS5 SPARQL Syntax & Intuition 11 JUN DS3 RDFS – Semantics 16 APR DS2 SPARQL – Semantics 14 JUN DS2 RDFS Rule-based Reasoning 16 APR DS3 SPARQL Algebra 14 JUN DS3 Tutorial 1 19 APR DS5 SPARQL 1.1 14 JUN DS5 OWL – Syntax & Intuition 26 APR DS5 Tutorial 6 18 JUN DS2 Tutorial 2 3 MAY DS5 SPARQL Entailment 18 JUN DS3 OWL – Syntax & Intuition ctd. 7 MAY DS2 Ontology Editing 2 JUL DS2 OWL & Description Logics 7 MAY DS3 Ontology Engineering 2 JUL DS3 OWL 2 10 MAY DS5 Tutorial 7 9 JUL DS2 Tutorial 3 14 MAY DS2 Linked Data 9 JUL DS3 Tableau I 14 MAY DS3 Applications 12 JUL DS5 Tableau II 17 MAY DS5 Test Exam 16 JUL DS2 Tutorial 4 7 JUN DS2 Test Exam Evaluation 16 JUL DS3 Hypertableau I 7 JUN DS3 Q&A Session 19 JUL DS5 TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 2 von 24

Ontology Engineering TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 3 von 24

Ontology Engineering TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 4 von 24

Agenda • How can we work with an ontology in a programming environment? • How can we work with an ontology via a graphical interface? TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 5 von 24

Agenda • How can we work with an ontology in a programming environment? • How can we work with an ontology via a graphical interface? TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 6 von 24

Implementing OWL What does “implementing OWL ” mean? • modeling – provision of data structures representing an ontology • parsing – conversion of the textual representation of an ontology (e.g. in RDF/XML) into corresponding data structures • serialization – conversion of the internal data structure into the textual representation • manipulation – provision of methods for manipulating the data structures or creating new ontological elements • inferencing – taking into account the OWL formal semantics TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 7 von 24

Aspects of an Implementation • identify functionalities and responsibilities • representation – syntax vs. data model – interface vs. implementation – locality of information • parsing/serialization – abstraction from the concrete representation (e.g. as triples) • manipulation – granularity – dependencies – intention of the user – strategies • inferencing – distinguishing explicit and implicit consequences – external imnplementations TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 8 von 24

What is an Ontology (from the Implementer’s Perspective) • a certain syntactic representation? • facts that are represented via the syntactic representation? • information that is entailed by the represented facts? TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 9 von 24

What is an Ontology (from the Implementer’s Perspective) • a certain syntactic representation? • facts that are represented via the syntactic representation? • information that is entailed by the represented facts? • these questions are important when working with OWL • e.g.: which answers to expect when querying these data structures TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 10 von 24

The OWL API • the OWL API is such an implementation of OWL • meant for OWL DL • an ontology is represented by a set of axioms, which model information about classes, roles and individuals • when is a class or role “in” an ontology? – not precisely defined in OWL spec – somewhat depending on the implementation TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 11 von 24

Data Structures of the OWL API • provides data structures forrepresenting ontologies • auxiliary classes for – creating, – manipulating, – parsing, – rendering, and – inferencing over these structures • the basic data structures represent objects of an ontology and are organized according to the functional syntax of OWL of OWL TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 12 von 24

The Ontology Object • an OWLOntology object represents an ontology • contains a set of axioms • “knows” what classes and roles are used in the ontology • may have an IRI as Name/ID and may be loaded from IRIs – the ontology’s ID does not have to coincide with the IRI the ontology is loaded from – same holds for imported ontologies TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 13 von 24

Loading an Ontology OWLOntologyManager manager =OWLManager.createOWLOntologyManager(); OWLOntologyIRIMapper aMapper =new AutoIRIMapper(new File("/ont/"), false); manager.addIRIMapper(aMapper); IRI rem=IRI.create("http://www.ex.org/ex.owl"); IRI loc=IRI.create("file:/ont/ex.owl"); OWLOntologyIRIMapper sMapper =new SimpleIRIMapper(rem, loc); manager.addIRIMapper(sMapper); OWLOntology ont =m.loadOntologyFromOntologyDocument( IRI.create("file:/ont/ont.owl")); TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 14 von 24

Objects for Entities • the class OWLClass represents an (OWL) class • does not contain information about which axioms use it • axioms referring classes belong to an OWLOntology object • methods in OWLClass allow acessing information about the class in the context of the ontology (convenience methods) • non-atomic classes use the interface OWLClassExpression • analogous for properties: OWLObjectProperty and OWLDataProperty TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 15 von 24

Adding Axioms OWLDataFactory df=m.getOWLDataFactory(); OWLClass student=df.getOWLClass(IRI.create("...")); OWLClass course=df.getOWLClass(IRI.create("...")); OWLObjectProperty takes =df.getOWLObjectProperty(IRI.create("...")); OWLClassExpression sup =df.getOWLObjectSomeValuesFrom(takes, course); OWLAxiom ax=df.getOWLSubClassOfAxiom(student, sup); manager.addAxiom(ont, axiom); TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 16 von 24

Using a Reasoner • the class OWLReasoner is an interface, that is implemented by most of the reasoners • contains methods, to interact with the reasoner import org.semanticweb.owlapi.reasoner.OWLReasoner; import org.semanticweb.HermiT.Configuration; import org.semanticweb.HermiT.Reasoner.ReasonerFactory; ... OWLReasonerFactory reasonerFactory =new Reasoner.ReasonerFactory(); // HermiT factory OWLReasoner reasoner =reasonerFactory.createReasoner(ont); System.out.println(reasoner.isConsistent()); System.out.println(reasoner.isSatisfiable(student)); hermit.precomputeInferences( InferenceType.CLASS_HIERARCHY); TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 17 von 24

Visitor patterns in the OWL API • visitor : Encapsulate an operation that is to be performed on the elements of an object structure as an object. The visitor pattern allows you to define a new operation without changing the classes of the targeted data structure. [Design Patterns, Gamma et al.] • allows for diverse operations over the data structures without overloading them with application-specific code • good as long as the data structures don’t change • changes in the OWL standards (very) rare TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 18 von 24

Example ClassExpressionVisitor • visitors implement the interface OWLClassExpressionVisitor which needs a methof visit() per OWL class object • all implementations of the interface OWLClassExpression must implement a method accept(OWLClassExpressionVisitor visitor) public interface OWLClassExpressionVisitor { void visit(OWLObjectUnionOf ce); ... } public interface OWLObjectUnionOf extends OWLClassExpression { public void accept(OWLClassExpressionVisitor v) { visitor.visit(this); } ... } TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 19 von 24

Applications for Visitors allow for working with arbitrary OWL objects, for instance • conversion of classes or axioms into NNF • simplification of axioms • serialization into diverse formats TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 20 von 24

Agenda • How can we work with an ontology in a programming environment? • How to work with an ontology via a graphical user interface? TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 21 von 24

Ontology Editing with Protege • load mad-cow-start.owl in Protege • familiarize yourself with the ontology TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 22 von 24

Ontology Editing mit Protege given the following text, extend the ontology by the concepts pet+owner, dog+owner and vegetarian: Pet owners are persons who own a pet. Dog owners are special pet owners in that they own a dog. Vegetarians are animals that do not eat other animals and neither do they eat parts of other animals. TU Dresden, July 2 Foundations of Semantic Web Technologies Folie 23 von 24