The possibilities in RDFS to define RDF vocabularies are, with respect
their limited expressiveness. That was a conscious decision
the design of RDFS. The language should be simple and understandable
be. The OWL, however, is a language for describing ontologies
and thus more expressive than RDFS. Due to the OWL, such as
be involved in RDFS, RDF knowledge bases. The language is a
Extension of RDFS with a larger number of functions to
to define class hierarchies and property restrictions.
These include relations between classes, cardinality (eg
"Exactly one"), equality, the more pronounced ways of typing
Properties and property characteristics (eg symmetry).
To compromise between expressiveness and efficient reasoning
to enable OWL has been divided in three languages. These three
Languages have different characteristics:
1st OWL-Lite is a limited form of OWL. It is optimized for
easy to understand and easy to implement to be. At
OWL-Lite are excluded disjoint expressions. Cardinality
be supported, but only the values
allowed 0 and 1. The disadvantage of the simple '' design of OWL-Lite
are limitations in the expressiveness.
2nd The OWL-DL includes all the language features of OWL-Full, but
partially limited in their use are. OWL-DL
designed for users who require maximum expressiveness
and at the same time predictability (all conclusions are guaranteed predictable)
and decidability (all computations terminate in
finite time) are dependent.
3rd OWL Full supports all language features of OWL and allows
an arbitrary combination of these characteristics with RDF and RDFS.
This, however, the flexibility has its price, at the expense of
Decidability of reasoning goes.
The term "DL" in OWL-DL stands for Description Logics. This
constitute the formal basis for OWL. Description logics are a
Family of languages belonging to the used for knowledge representation
can be. Its origin lies in the frames-based systems
Knowledge representation. The term was originally proposed by Marvin Minsky
introduced.
A frame is a structure to represent a concept or a
Situation. Frames can be covered with different types of information
be. These include, for example, information on the definition, description
and use of the frame. The description logics are important
Descendants Repräsentationsformalismen frame based, the declarative
Percentage of frames, using a logic-based semantics
underwriting. Then they left parts of the first order predicate logic
based and are decidable, they can hit on conclusions
can be (that is already present in New knowledge gained
be). In description logics, it is possible, a concept hierarchy
build from atomic concepts and attributes. The attributes are
it usually called roles. The intended meaning of
atomic concepts may be replaced by other concepts and role restrictions
be established.
A simple example:
Woman = person and female being
Parent = person with a child
Grandmother = woman with a child, the parent is.
One of the main tasks of reasoning in this context
The question is whether all instances of a concept necessarily
Other instances of the concept are. An answer to this question is to
be found in the light of all definitions. Thus, for
Example, the "grandmother" as a "parent classified" because each "grandmother"
By definition, the same as "parent" is. Similar to the task
the classification is the task of reviewing instances. This
the answer will be determined on the question of whether a given object is an instance
of the previously specified concept. Description logics have
as shown, is also related to object-oriented concepts in
Programming.
