Database Reference
In-Depth Information
properties and represent relationships between resources, adding semantics
to the terms in a vocabulary. Intuitively, RDF allows us to describe instances,
while RDFS adds schema information to those instances. A comprehensive
study of the formal semantics of RDFS is beyond the scope of the topic, but
we provide below the basic concepts we will use in the next sections.
Among the many terms in the RDFS vocabulary, the fragment which
represents the essential features of RDF is the subset composed of the
following terms:
rdf:type
,
rdf:Class
,
rdfs:Resource
,
rdfs:Property
,
rdfs:range
,
rdfs:domain
,
rdfs:subClassOf
,and
rdfs:subPropertyOf
. For example, a triple
Employee rdf:type Class
tells that
Employee
is a class that aggregates objects
of the same kind, in this case employees (we leave syntactic issues to be
presented later, since actually all resources must be defined using IRIs).
The triple
Davolio rdf:type Employee
tells that Davolio is a member of the
class
Employee
.Theterm
rdfs:Resource
denotes the class of all resources, and
rdf:Property
the class of all properties. Importantly, class membership is not
exclusive, since a resource may belong to several different classes. Elements
belonging to the class
rdf:Property
represent relationships between resources,
used in the predicate part of RDF triples. For example,
hasSalary
can be
defined as a property of an employee using the statement
hasSalary rdf:type
rdf:Property
. The predicate
rdfs:subClassOf
allows us to define generalization
relationships between classes. For example, the triple
TemporaryEmployee
rdf:subClassOf Employee
tells that every temporary employee is also an
employee. Analogously, the predicate
rdfs:subPropertyOf
allows us to define
generalization relationships between properties. For example,
hasLowSalary
rdfs:subPropertyof hasSalary
indicates a subproperty to describe employees
with low salaries. A rule system can be defined using these and other
predicates, thus allowing to infer knowledge from an RDF graph. The
RDF Semantics
4
specification defines a precise semantics and corresponding
complete systems of inference rules for RDF and RDFS. Finally, let us remark
that, in general, triples representing schema and instance data coexist in RDF
data sets.
14.1.2 RDF Serializations
An RDF graph is a collection of triples given in any order, which suggests
many ways of serialization. Two widely used notations are
RDF/XML
,
5
which defines an XML syntax for RDF, and
Turtle
,
6
which provides a simple
way of representing RDF triples.
Search WWH ::
Custom Search