Information Technology Reference
In-Depth Information
Class(Person partial
restriction (hasChild allValuesFrom(Person)))
Class(Parent complete
Person
restriction (hasChild someValuesFrom(Person)))
ObjectProperty(hasChild)
Individual (John type(Person)
value(hasChild Mary))
Listing 3.1.
OWL example using abstract syntax
the class Person. Although this inference seems to be intended in this scenario,
it cannot be assumed that this kind of inference is intended in general [31].
3.4 Rules for the Semantic Web
From the early days of the Semantic Web, rules have been seen as an important
paradigm for representing and reasoning with knowledge on the Semantic
Web.
6
However, at the time of writing, activities have just gotten underway
towards standardization of a rules language for the Semantic Web.
7
With ontologies, one can express knowledge about classes, class hierarchies,
properties, etc. Rules have a complementary expressiveness: with rules, one
can express knowledge in the form “if
A
then
B
”. An example which is often
used to motivate the use of rules is the “uncle” example, which says that “the
brother of a person's parent is that person's uncle”:
person
(?
x
)
∧
parent
(?
x,
?
y
)
∧
brother
(?
y,
?
z
)
⇒
uncle
(?
x,
?
z
).
3.4.1 SWRL
SWRL [64] is an extension of OWL DL which adds the expressive power of
rules (without negation) to OWL; the above “uncle” example can be expressed
in SWRL.
The basic SWRL constructs are Horn-like rules. However, whereas Horn
rules have a conjunction of atomic formulas in the antecedent (body) of the
rule and a single atomic formula in the consequent (head) of the rule, SWRL
allows any OWL class description, property, or individual assertion in both
the body and the head of the rule. In this way, SWRL diverges from tradi-
tional rules systems, which are based on Logic Programming or Deductive
Databases.
6
http://www.ruleml.org
.
7
http://www.w3.org/2005/rules
.
Search WWH ::
Custom Search