=Paper=
{{Paper
|id=None
|storemode=property
|title=Ontology of Ontology Patterns as Linked Data Integration Tool
|pdfUrl=https://ceur-ws.org/Vol-631/paper7.pdf
|volume=Vol-631
}}
==Ontology of Ontology Patterns as Linked Data Integration Tool ==
https://ceur-ws.org/Vol-631/paper7.pdf
Ontology of Ontology Patterns as Linked Data
Integration Tool
Miroslav Vacura and Vojtěch Svátek
Faculty of Informatics and Statistics,
University of Economics
W. Churchill Sq.4, 130 67 Prague 3,
Czech Republic
vacuram|svatek@vse.cz
Abstract. The paper present preview of ontology of ontology design
patterns and transformation patterns being developed as support tool
for emerging ontology design techniques and methodologies.
The Linked Data initiative was started by Tim Berners-Lee as an architec-
tural vision for the Semantic Web. It explores the idea of Semantic Web as
puting emphasis on making links so both people and machines can explore the
interconnected web of data [1].
knows
Relation-type
Person Person
FOAF T1 Person (n-ary relator) Person
T2
Role
Relation-type Role
Person (n-ary relator) Person T3 Relation-type
Event Person (n-ary relator) Person
Event
Time-interval Event
Fig. 1. FOAF Transformations
As an use-case we choose the FOAF project and it’s ‘knows’ relation. Since
2004 there was more than 1 million FOAF documents and 79% of them utilized
the knows property [2]. The typical needs of ontology engineer working on top
of some Linked Data source comprises of transformation or aligning data to
some more complex ontology either newly designed or already existing. This is
also case of knows relation that in its FOAF implementation is very simple and
doesn’t allow expressing more complex relations among individuals.
In the case of newly designed ontology use of ontology design patterns (ODPs)
proved the most effective and the least time consuming way of doing it. In con-
text of our use-case we can think of several iterations of ODPs that represent
more or less complex or expanded view on ‘knows’ relation as is depicted on
Fig. 1. These ODPs can be connected together using predefined transformation
patterns (OPPL), It seems to us that having library of such predefined pattern
transformations at hand could make such design tasks easier and much faster.
Our ongoing work proposes development of such library on top of existing por-
tal OntologyDesignPatterns.org, but 1) in form of ontology, 2) with explicitly
stated relations, 3) that are formally defined and 4) with appropriate transforma-
tions (OPPL) between related patterns, that enable automatic transformation
from one pattern to another. We also focus on providing more fine-grained anal-
ysis of relations (like specialization/generalization) between ontology ODPs.
SpecialisationOf
Ontology Patterns (OP)
Logical ODP
SourceODP
Content ODP ODP OTP
TargetODP
Fig. 2. Design Patterns Transformations Ontology
Such a library would could be easily integrated with methodologies like Ex-
treme Design (XD) [3] and respective development tools like NeON or Protegé.
This extended abstracts presents early preview of architecture of ontology being
developed on Fig. 2. For additional information see http://keg.vse.cz.
Acknowledgments
This work has been partially partially supported by the IGS 4/2010 and by
the CSF grant no. P202/10/1825 (PatOMat - Automation of Ontology Pattern
Detection and Exploitation).
References
1. Tim Berners-Lee. LinkedData. http://www.w3.org/DesignIssues/LinkedData.
html, 2009.
2. Li Ding, Lina Zhou, Tim Finin, and Anupam Joshi. How the semantic web is being
used: An analysis of foaf documents. In Proc. of the 38th ICSS, 2005.
3. Valentina Presutti et al. eXtreme Design with Content Ontology Design Patterns.
In Eva Blomqvist et. al., editor, Proc. of the WOP 2009, coll. with ISWC-2009,
volume 516. CEUR Workshop Proceedings, 2009.