The Semantic Web - Enabling Semantic Web Services: The Web Service Modeling Ontology - page 34

classes, instances, and attributes depending on the syntactic context, thereby
allowing certain kinds of metastatements. For example, we might define an
attribute parent , which is not related to the class parent :
person[parent =>> person].
Both SWRL and F-Logic have been proposed as rules languages for the
Semantic Web. The main difference between the two proposals is that in
SWRL, the rules language is seen as an extension of the ontology language
OWL DL, whereas in the F-Logic (programming) proposal, ontologies are
modeled using rules.
3.5 Summary
In this chapter we have seen that Semantic Web technologies have the poten-
tial to overcome many of the limitations of the current Web by introducing
machine-processable annotations and formal languages for expressing knowl-
edge on the Web. Ontologies play a key role in enabling communication and
interoperation on the Web because they represent consensual terminologies
which can be used for automated processing.
RDF [75] is the language recommendation from the W3C for writing
machine-processable annotations on the Web. RDFS [20] and OWL [32] are
the language recommendations for writing ontologies on the Web. RDFS is an
inexpressive ontology language which offers the possibility to define classes,
class hierarchies, properties, property hierarchies, and domain and range re-
strictions. OWL extends RDFS towards a more expressive ontology language.
Some examples of the features of OWL are transitive, symmetric, and inverse
properties; local restrictions on properties; cardinality and existential property
restrictions; and boolean combination of classes.
Rules can be seen as a knowledge representation paradigm which is or-
thogonal to that of ontologies, yet can also be combined with ontologies
(e.g. SWRL [64]). However, it has also been argued that rules should be used
as the basis for an ontology language [69].