Figure 2‑9 Semantic Web in the years of the schedule of components within the body.
What is ontology?
Ontology is the key technology to be used to perform the semantic web.
The term ontology as a science of being in philosophy
defined. Ontology of the twentieth century
At the end, it was used extensively in the field of artificial intelligence. Ontology in a certain area
sharing and reuse of information
Providing “conceptualizations formal and
defined as a clear definition ”and
used. Recently ontologies
Intelligent Systems, Electronic Commerce,
on natural language processing and information management
is becoming widespread.
Ontology in computer science
The definition of accepted Gruber (1993);
"Clearness of conceptualization"
as expressed in the form of. Conceptualization, abstract model in a certain design stage
means creation. Maedche and Staab
(2001).
clearly defined in a way that does not leave
must.
Another definition of ontology given in Maedche and Staab (2001) is what ontology
explains that it should include and what models containing should be considered ontology.
The difference of this definition from others; The word here
It is the web ontologies of ontologies.
Each term is identified and the terms found in ontology are finite. These;
The meanings of terms, between terms
relationships, synonyms between terms
And it includes sequential relationships. In conclusion,
At least in order to have an ontology of a set of concepts, at least have the following qualifications.
must be:
• Terim repertoire must be finite and expandable,
• Relations between classes and terms
that it is clearly understood by the factors,
• OWL's use of the class structure.
In addition to three basic qualifications, in order to increase the expression power of ontologies and get qualified inquiry results,
• Don't contain samples,
• Having class qualities,
• Value comparison criteria,
• Logical comparison criteria,
The properties can also be added to the ontology qualities given above. With the inclusion of these logical structures; ontologies, complex
can model the information and on this information
They come to a level that can produce results.
Web ontology language
Web Ontology Language (OWL), ontologies
A used to define and diversify
Language (Wikipedia, 2007). OWL, not only showing the content of information to people, it
to be processed by computers next to it
designed to. OWL, XML, RDF and
Supported by RDF Schema (RDF-S)
Web content is better by computers
aims to ensure that it can be interpreted. Languages like OWL, OIL and Daml+Oil
It is based and now by the W3C institution
It is a recommended language. OWL, Drop+OIL
Web is a research -based version of ontology (Wikipedia, 2007).
Figure 2‑10 The schematic structure of OWL languages
An example of the use of the OWL language as RDF is given in Figure 2‑10.
Connection Between Terms RDF syntax
is done with. SYLING RULES INFORMATION
Designed to make sense. This is
We can explain it with an example. A person
Apple in the sentence "I use Apple"
To define the word as RDF
Apple word; manufacturer, type, model,
When it defines the operating system, it implies meaning on the word Apple. Computer Apple
When he reads the word, he will realize that it is a defined computer. This
Figure 2‑10 is shown with the language of ontology.
For a single purpose, the information defined by OWL
unthinkable. This information can be defined or expanded by other ontologies.
Not misinterpretation of OWL documents and processing by computers
There is a need for a proper syntax.
Terminology to be used in OWL documents
It is the extended version of RDF files (OWL,
2007).
OWL has three sub-languages: OWL Lite, OWL DL and OWL full as shown in Figure 2-11.
These three languages are used by users
They are specially designed (OWL,
2007). The sub -languages of the OWL language have been designed to cover each other. Figure 2‑11
This structure is shown.
Figure 2‑11 Language Structure that constitutes OWL language
OWL FULL: RDF and RDF (S) uses all components. All conveniences provided by RDF
provides. However, it does not support computability.
For example, the class described in the RDF can express a community as well as defining an individual
It may also be a class. OWL Full is more important than the calculation information of the definition information
can be used in cases. OWL FULL
In some cases, automatic logic can be impossible. OWL full language with RDF
fully compatible. RDF documents
Its structure is also OWL Full Documents. One of the disadvantages OWL Full Structures
is not fully determined. Therefore, concepts
it cannot make a full and effective logic between it.
For example, Ahmet Bey get a printer from the techno market that sells a computer material.
Logically, Ahmet Bey's printer like cartridge, toner or ribon for the printer in the future
can buy one of the consumables
Potential customer can be concluded. The type of printer he bought is certain
Since it is not, the consumables it will use for the printer
The material can be very diverse. The interest of interest in the logical expression created in this way
As a result of its weakness, the company can decide which consumables to invest in the material
There are multiple options.
As a result, such logical inferences are the only
It does not create an effective logic in achieving a result.
OWL DL: It is a sub -language of OWL Full. OWL
and used by adding constraints to RDF -forming structures. These constraints define logic
Provides connection. OWL DL provides convenience in defining the meanings of terms. Reason by making automatic calculations possible
Make it easier to execute. For example, Ahmet Bey
Let's assume that you buy a laser printer from the technology market. If we make logical execution; If Ahmet Bey bought a laser printer
For the future for the printer
will take a toner. A logically laser printer, a person with a laser printer should get a toner for the laser printer.
logic is being carried out.
OWL Lite: uses a single part of the OWL language. Uses the restrictions of OWL DL. OWL
Lite can only be defined as the upper class. Decision structures are used in the upper classes. Between classes
There are equality and subclass relations. OWL Lite
It sends two values by evaluating the logical consequences. These values are 0 and 1. Especially
In case of signs; If I have a interest
1 is accepted. Other cases are considered 0.
For example, the processor of a computer,
Motherboard, RAM memory server
If it has a feature, it sends 1 value, and in other cases 0 value is sent.
How to create ontology?
To create ontology, primarily an area
should be determined. There should be a finite number of terms on this area. The classes of these terms and
sub -classes should be determined. Individuals of classes
should be defined. Class definitions may be in a reference class. But the lower classes are reference
cannot be a class. Sub -class definitions usually
Used to identify more private individuals.
For example, the class when creating information ontology
as a sub -class as a sub -class object -oriented programming languages terms.
We can write (object, class, function).
Figure 2‑12 Class structure representation
A very important definition in creating ontology is to indicate the characteristics of the classes. Each class has its own specific feature. While making characteristics to classes
Data types and object properties can be specified. Fig.
is determined. This upper class is called domain.
The connection address is called range. In addition, the concept of domain is between programming types
Used to control marlling. Ranking
The concept determines the connection points of the term. At the same time the concept of domain from classes
It can be used in information inference. Your every class
There is a domain area. This Domain is a URI
can be the address. Semantic in creating ontology
Bonds are reflected in the classes. This reflection belonging
performs the action. The individual specifies his belonging to the class only with an action. These actions
with ownership suffix.
Simply exemplary an example ontology scheme
And the creation of ontology language:
Theoretical to create an ontology so far
as we explained. We will try to show how to create an ontology. The area where we will create an example ontology
Informatics terms. After the field is determined
The classes of the terms of informatics and these classes
Let's group the terms according to their interests. We
As class:
Network Terms
(DNS, IP, RJ45,…)
• Software Terms
(Object, function, condition,…)
• Terms of the frost
(Network Card, Processor, Memory,…)
Let's define three classes, including. After determining these definitions schematically
It is time to use ontology creation tools
income. Protégé as a means of creating ontology
3.3.1 version can be used. Protégé program,
It is distributed free of charge. After determining the classes of terms; if necessary
We need to determine the subclasses of classes.
For example, the subclasses of hardware terms
MIB TERMS
(Cape memory, CPU, RAM memory,…)
• Memory Terms (Static Memory,
Dynamic memory, swap memory,…)
• I/O Terms
(Printer, browser, monitor,…)
We can divide it into lower classes.
After the lower classes are defined, the terms
Features and relationships are determined. Later
The actions of these relationships are defined. For example;
“Processor uses memory. The memory of the graphics card
has. Uses C ++ memory. Game software
Dynamic memory and the memory of the graphics card
We can make connections on the term memory term as a term. Once the definitions and relationships are determined, ontology is created using ontology language.
starts. Classes as shown in Figure 2‑3
and sub -classes were determined. Subclasses
Types and connections of terms as code
given.
Connections of Terms In the code
Is written in the form of URI. Connection shape
We can examine it by looking at the sample code (Code: 2.1).
Figure 2‑13 Creation Classes and Sub -classes
Code: 2-1
<? xml version = ”1.0”?>
<RDF: RDF XML: Base = ”http://ww.owlontologies.com/unnamed.
OWL ”xmlns =” http: // www.
Owl-ontologies.com/unnamed.
OWL#”XMLNS: OWL =” http: //
www.w3.org/2002/07/owl# ”
XMLNS: RDF = ”http: // www.
w3.org/1999/02/22-rdf-syntaxns# ”XMLNS: RDFS =” http: // www.
w3.org/2000/01/RDF-schema# "
xmlns: xsd = ”http: //
www.w3.org/2001/xmlschema#”>
<OWL: Ontology RDF: About = ””/>
<OWL: Class
RDF: ID = ”Hardimathers”/>
<OWL: Class
RDF: ID = ”memory for”>
<rdfs: subclassof
“RDF: Resource
= ”#Hardware
</uwl: class>
<OWL: Class
RDF: ID = ”Networkterimleri”/>
<OWL: Class
RDF: ID = ”Databasers”>
<rdfs: subclassof>
<OWL: Class
RDF: ID = ”softwareterimations”/>
</dfs: subclassof>
</uwl: class>
<OWL: Class
RDF: ID = ”Algorithmaterials”>
<rdfs: subclassof
rdf: resource = ”#software
Terms ”/>
</uwl: class>
…
<Webterims RDF: ID = ”ipdns”/>
<Webterims
RDF: ID = ”ipipx-SPX”/>
<database
RDF: ID = ”Join”/>
<database
RDF: ID = ”innergoin”/>
<database
RDF: ID = ”Function”/>
…
<rdf: description
RDF: ID = ”Knowledge_
Instance_6 ”>
<RDF: Type> CLS (: Instanceannotionion,
FRAMEID (0: 1020 0))
<programmatic
RDF: ID = ”Java”/>
<database
RDF: ID = ”Leftjoin”/>
<database
RDF: ID = ”update”/>
<Algorithmaterials
RDF: ID = ”WHile”/>
<programmatic
RDF: ID = ”C”/>
<programmatic
RDF: ID = ”OWL”/>
</ rdf: rdf>
Code: When we examine 2.1, Parenthesis 1 XML
Name Space Definitions, 2.
Place where parenthesis classes are defined, 3rd parenthesis
where lower classes are defined. 4. It can be examined as the place where parentheses are defined.
Conclusion
Ontologies work conceptually with the logic of algorithm. From the structure of the algorithm
Unlike the connection between objects
It is determined by action. These actions are in the form of terms in ontology. In addition, the sentence that makes the action; Subject, object
and in the form of predicate. For example,
Cats like fish. Old people are usually a
He has a cat. So old people buy fish for their cats.Nowadays, ontology creation is usually done by computer software developers.
The creation of ontology shortens the definition and results of the works to be done. In the structural transformations of institutions with the spread of ontology in the future,
In task definitions and business shares
The decision -making mechanisms of institutions are more
It is expected to work fast.
According to Cimiano (2006), the most popular ontologies today, Wordnet ontology, Wiki ontology, Eclips ontology, Medical Ontology,
Gene ontology. These ontologies are
Common feature; to be accepted as standard
and in accordance with open source code principles
distributed.
