OWL: Web Ontology Language

Download OWL: Web Ontology Language

Post on 23-Feb-2016




0 download

Embed Size (px)


OWL: Web Ontology Language. Slides are from Grigoris Antoniou, Frank van Harmelen, A Semantic Web Primer. Limitations of RDF Schema. Equivalence of classes - PowerPoint PPT Presentation


PowerPoint Presentation

1OWL: Web Ontology Language

Slides are from Grigoris Antoniou, Frank van Harmelen, A Semantic Web Primer

12Limitations of RDF SchemaEquivalence of classesTwo classes, same concept - people use different words to represent the same thing. It would be very useful to be able to state "this class is equivalent to this second class".E.g. One person may create an ontology with a class called "Airplane". Another person may create an ontology with a class called "Plane". It would be useful to be able to indicate that the two classes are equivalent.Boolean combinations of classesSometimes we wish to build new classes by combining other classes using union, intersection, and complement.E.g. Person is the union of the classes Male and Female persons.Disjointness of classesSometimes we wish to say that classes are disjoint e.g. male and female are disjoint classes

23Limitations of RDF Schema (continued)Cardinality constraints - oftentimes it is useful to indicate the allowable number of occurrences of a propertyE.g. We would like to be able to express that a River has only "one" officialLength property. E.g. We would like to be able to express that an Ocean has one maxDepth.E.g. a person has exactly two parents, a course is taught by at least one lecturerSpecial characteristics of properties Transitive property (like greater than)Unique property (like is mother of)A property is the inverse of another property (like eats and is eaten by)

34RDF Schemas: Building Block to More Expressive Ontology LanguagesRDF Schema was designed to be extended. The ontology languagesall use RDF Schema's basic notions of Class, Property, domain, and range.

45What is OWL?OWL stands for Web Ontology Language OWL is built on top of RDF OWL is written in XML OWL is for processing information on the web OWL was designed to be interpreted by computers OWL was not designed for being read by people OWL is a web standard OWL has three sublanguages (version 2004)OWL Lite, DL, FullOWL 2 defines several profiles with different applications in mindThe following slides are from Grigoris Antoniou, Frank van Harmelen, A Semantic Web Primer56Tradeoff between Expressive Power and Efficient Reasoning SupportThe richer the language is, the more inefficient the reasoning support becomesSometimes it crosses the border of non-computabilityWe need a compromise:A language supported by reasonably efficient reasoners A language that can express large classes of ontologies and knowledge.

67OWL SublanguagesOWL Lite Simple constraints and classification hierarchyOWL DL (includes OWL Lite) Maximal expressiveness while retaining computational completeness (all conclusions are computable) OWL Full (includes OWL DL) Maximal expressivenessUnlikely any reasoning software to support OWL Full.

OWL LiteOWL DLOWL Full7It uses all the OWL languages primitivesIt allows the combination of these primitives in arbitrary ways with RDF and RDF SchemaOWL Full is fully upward-compatible with RDF, both syntactically and semanticallyOWL Full is so powerful that it is undecidableNo complete (or efficient) reasoning support

OWL DL (Description Logic) is a sublanguage of OWL Full that restricts application of the constructors from OWL and RDFApplication of OWLs constructors to each other is disallowedTherefore it corresponds to a well studied description logicOWL DL permits efficient reasoning supportBut we lose full compatibility with RDF: Not every RDF document is a legal OWL DL document. Every legal OWL DL document is a legal RDF document.

An even further restriction limits OWL DL to a subset of the language constructorsE.g., OWL Lite excludes enumerated classes, disjointness statements, and arbitrary cardinality.The advantage of this is a language that is easier tograsp, for usersimplement, for tool buildersThe disadvantage is restricted expressivity

Upward compatibility:Every legal OWL Lite ontology is a legal OWL DL ontologyEvery legal OWL DL ontology is a legal OWL Full ontology

Every valid OWL Lite conclusion is a valid OWL DL conclusionEvery valid OWL DL conclusion is a valid OWL Full conclusion

OWL 2 and sublanguages (profiles)OWL DL is decidable, but the performance is poor for non-trivial knowledge baseOWL 2 (2009) gives several profiles of OWL languageEach is intended for different users/applicationsInclude EL, QL, RLOWL 2 ELHas polynomial time to determine consistency of an ontologyA restriction on DLDisallowed class descriptions CardinalityAll-values-fromUnion-of, disjoint union, complement-ofDisallowed property descriptionInverse, disjointFunctional, symmetric,

8OWL 2 QL Efficient querySimilar to Entity-relationship modelingGood to model information contained in existing databaseEnable queries in logspace wrt knowledge base size

910Reasoning About Knowledge in Ontology LanguagesClass membership E.g., If x is an instance of a class C, and C is a subclass of D, then we can infer that x is an instance of D.Subclass and Equivalence of classes E.g., If class A is equivalent to class B, and class B is equivalent to class C, then A is equivalent to C, too.ConsistencyE.g., X is an instance of classes A and B, but A and B are disjoint.This is an indication of an error in the ontology.

1011Uses for ReasoningReasoning support is important forchecking the consistency of the ontology and the knowledgechecking for unintended relationships between classesautomatically classifying instances in classesChecks like the preceding ones are valuable for designing large ontologies, where multiple authors are involvedintegrating and sharing ontologies from various sources

1112Reasoning support in OWLSemantics is a prerequisite for reasoning supportFormal semantics and reasoning support are usually provided by mapping an ontology language to a known logical formalismusing automated reasoners that already exist for those formalismsOWL is (partially) mapped on a description logic, and makes use of reasoners such as FaCT and RACER Description logics are a subset of predicate logic for which efficient reasoning support is possible1213OWL Compatibility with RDF SchemaAll varieties of OWL use RDF for their syntaxInstances are declared as in RDF, using RDF descriptions and typing informationOWL constructors are specializations of their RDF counterparts Rdfs:ResourceOwl:DatatypePropertyowl:ClassRdfs:PropertyRdfs:ClassOwl:ObjectPropertyOWLRDF/RDFSXML Schema1314OWL Syntactic VarietiesOWL builds on RDF and uses RDFs XML-based syntaxOther syntactic forms for OWL have also been defined:An abstract syntax, that is much more compact and readable than the XML languagesA graphic syntax based on the conventions of UML

1415OWL XML/RDF Syntax: Header

An example OWL ontology University Ontology

owl:imports is a transitive property

1617OWL ClassClasses are defined using owl:Classowl:Class is a subclass of rdfs:Class

owl:equivalentClass defines equivalence of classes

Disjointness is defined using owl:disjointWith

owl:Thing is the most general class, which contains everythingowl:Nothing is the empty class1718PropertiesIn OWL there are two kinds of propertiesObject properties, which relate objects to other objectsE.g. is-TaughtBy, supervises

Data type properties, which relate objects to datatype valuesE.g. phone, age, etc.

OWLRDF/RDFSXML Schema1819Relations to other propertiesEquivalent property

Inverse property

1920Property restriction Value constraints: put constrains on the range of the property when applied to this particular class description.owl:allValuesFrom specifies universal quantification owl:hasValue specifies a specific value owl:someValuesFrom specifies existential quantificationCardinality constraints: the number of value a property can take.

It is a way to define classes

2021owl:allValuesFromDefine the class of persons whose parents are both physicians. { x | (x, y) hasParent implies y Physician }Example:Suppose: David hasParent Bill who is a Physician, and David hasParent Rose who is a PhycisianConclusion: David belongs to the class defined above.

Semanticsrestriction(p allValuesFrom(r)) : {x | (x,y) p implies y r}

2122allValuesFromFirstYearCourse is a course that must be taught by professors

{ x | (x,y) isTaughtBy implies y Professor }

23owl:someValuesFromDefine the class of persons who has at least one physician parent.

{ x | (x, y) hasParent y Physician}

Semanticsrestriction(p someValuesFrom(e)) : {x | (x,y) p y e} 2324someValuesFromAcademicStaff must teach at least one under course.

{ x | (x, y) Teaches y undergraduateCourse}


restriction(p value(i)), for i an individual ID {x | (x, i) p}2526Cardinality restrictionsWe can specify minimum and maximum number using owl:minCardinality and owl:maxCardinality It is possible to specify a precise number by using the same minimum and maximum numberFor convenience, OWL offers also owl:cardinalityE.g. Every course is taught by at least someone:


2627Example: Book Ontology



1 28Special propertiesowl:TransitiveProperty (transitive property) E.g. has better grade than, is ancestor ofowl:SymmetricProperty (symmetry)E.g. has same grade as, is sibling ofowl:FunctionalProperty defines a property that has at most one value for each objectE.g. age, height, directSupervisorowl:InverseFunctionalProperty defines a property for which two different objects cannot have the same value

2829Boolean Combinations: complementOfWe can combine classes u


View more >