RDFS

The Resource Description Framework Schema (RDFS or RDF Schema) is a semantic extension of RDF. RDF Schema complements RDF by providing a schema language which allows the definition of RDF vocabularies. RDF Schema is RDF's vocabulary which allows to define concepts and their relationships. It provides mechanisms for describing groups of related resources and the relationships between these resources - simple ontologies, models about concepts and their properties.

RDF Schema vocabulary descriptions are written in RDF. These resources are used to determine characteristics of other resources. RDF Schema defines constructs for specifying which classes, predicates and individuals exist in the vocabulary. It specifies how concepts may relate to one another.

Classes and Individuals

RDF Schema deals with classes and individuals. Classes are resources to which other resources can belong. The resources (instances) which belong to a class are individuals.

Relationships between Classes and Individuals

RDF Schema deals with the relationships between classes and individuals and between classes. These relationships are rdf:type, rdfs:subClassOf.

The individual - class relationship is expressed with rdf:type. For instance,

[1] Peter rdf:type Doctor

encodes the information that Peter belongs to the class of Doctors.

The class - subclass relationship is expressed with rdfs:subClassOf. For instance,

[2] Doctor rdfs:subClassOf Person

encodes the information that the class of Doctors belongs to the class of Person.

Using the inference mechanics one can infer from these two statements that

[3] Peter rdf:type Person

Properties

Properties in RDF are the predicates connecting two entities (subject and object). In RDF Schema properties have domain and range.

The domain refers to the class of the subject of the property. The range refers to the class of the object of the property. For instance,

isBornIn rdfs:domain Person
isBornIn rdfs:range Location

denotes that Person is the domain of the property isBornIn and Location is the range of the property isBornIn. This means that the individuals in a subject position of the property isBornIn have to be of class Person, and the individuals in object position of the same property have to be of class Location.

This allows to infer from the statement:

Peter isBornIn Sofia

that Peter is a Person and Sofia is a Location.

Subproperties

Properties in RDFS have also subproperties, e.g. rdfs:subPropertyOf. This allows to define hierarchies of properties, and infer new knowledge with them.

There are many techniques through which the meaning of classes and properties can be described.

RDFS semantics

RDF Schema semantics supports 33 entailment rules. These entailment rules define the inference capabilities of RDFS. For instance, the rule below defines the relationship between an inividual and two classes. It is of the two classes, because it is declared to be a member of a class which is a subclass of the other class. The transitive nature of the subclass relation makes the inference possible.

if a rdf:type x and x rdfs:subClassOf y then a rdf:type y

Another example of an entailment rule pertains to the class of the range of a property. It states that if the range of a property is of a certain class, then the individual which is in the range of this property is of this class.

if P rdfs:range C and x P y then y rdf:type C

The Web Ontology Language (OWL) is an extension of RDF Schema. It makes richer models of knowledge about things possible, but at the cost of those models being more complex for a computer to process.