18-08-2012, 01:35 PM
A SURVEY ON ONTOLOGY CONSTRUCTION METHODOLOGIES
A SURVEY ON ONTOLOGY CONSTRUCTION METHODOLOGIES.pdf (Size: 90.44 KB / Downloads: 63)
Introduction
Ontology is a formal explicit specification of a shared conceptualization [1]. Gonzales [2] explores
the terminologies in the above definition and states that formal is an abstract model of portion of the
world; explicit specification means that the constructed ontology must be machine readable and
understandable; shared implies consensus of the community towards the ontology that have been built
and conceptualization is expressed in terms of the concepts and the properties of the ontology.
Ontology is also expressed as a formal representation of knowledge by a set of concepts within a
domain and the relationship between these concepts. Swartout et. al [19] divides ontology into two
categories: domain ontology and theory ontology. Domain ontology deals with the formal description of
the classes, the relationship between the classes, while theory ontology tends to be abstract and smaller.
It mainly deals with time, space, plans etc.
Advantages and Need for Ontology
Enable reuse of already existing domain knowledge instead of creating a new one.
• Make domain assumptions unambiguous
• Examine domain knowledge.
Ontology has gained much importance not only in the field of artificial intelligence but also in the fields
like information organization, natural language processing, information retrieval, knowledge
representation, knowledge acquisition. Ontologies are popular for the following reasons.
• Reusability- Instead of creating a new ontology the existing ontology can be reused.
• Formal community time- Shared viewpoint over a certain universe of discourse.
• Interoperability- Support communication and co-operation among systems
• Knowledge level validation and verification – checks for the completeness of domain discourse.
Components of Ontology
Ontology consists of four main components to represent a domain. They are:
i. Concept represents a set of entities within a domain.
ii. Relation specifies the interaction among concepts
iii. Instance indicates the concrete example of concepts within the domain
iv. Axioms denote a statement that is always true.
Let us take an example of a wine ontology and look at its components. The concepts of the wine
ontology are, “Winery, Wine, Wine descriptor, Wine color, etc”. The relationships are given as Winery
produces wine, wine haswinedescriptor wine descriptor. The instances of winecolor can be red, rose and
white. The axiom in this example is ‘a winery must produce atleast one type of wine’.