![](bg5.png)
sUniversity (University#1, name, PoitiersUniver-
sity) and (University#3, name, ToursUniversity).
Given that the idUniv property is computed as the
primary key of the University class, the loaded
triplets present a violation case of the uniqueness
constraint. Indeed, the two universities are dif-
ferent but they have the same identifier. In or-
der to detect such data, we propose the rule def-
inition to address the primary key violation for
each stored ontological class. The definition of
such a rule requires the creation of a rulebase, fol-
lowed by an index creation. This rule raises the
uniquenessconstraintviolation case by generating
a comment to prevent users about the data incon-
sistency existence as follows: (s, :comment, ’In-
consistency: violation of primary key constraint’)
where s presents the inconsistent instance. Once
the inconsistent data error is raised, the user will
decide if the data should be deleted or not.
4.2 Synthesis
In the most S D B , storage models are frozen. For ex-
ample, in Oracle 11g, the vertical representation is
used for the storage of ontological concepts and in-
stances referencing them. Therefore, our approach
can not be applied globally. This does not elimi-
nate the contributions offered by our methodology for
such databases. Indeed, based on the class dependen-
cies, the class’s types are identified and stored in the
ontology-based data models. The property functional
dependencies modeling allows to compute primary
keys for each ontological class and subsequently, to
store them in the S D B . Based on the basis of these
keys, the rules helping to detect a set of inconsistent
data may be defined. They ensure that the stored data
correspond to the boundaries of the modeled universe
and reduce the inconsistency and redundancy in the
S D B models.
5 CONCLUSIONS
This paper presents a complete methodology for de-
signing S D B covering the three main steps of tradi-
tional database design: conceptual, logical and physi-
cal. Ontology is the core of our design methodology.
Traditional definition of ontology is enriched by de-
pendencies between properties and classes. These de-
pendencies are exploited to generate consistent S D B .
They are used to identify the canonical concepts that
have to be stored in the database. The non-canonical
concepts are managed by relational views. A case
study showing the deployment of our S D B obtained
by our methodology in the semantic Oracle DBMS is
proposed.
Currently, we are developing a design tool to as-
sist designers during the S D B design process. Also,
we are working on incorporating optimization struc-
ture selection during the physical design.
REFERENCES
Borgida, A. and Weddell, G. E. (1997). Adding uniqueness
constraints to description logics (preliminary report).
In Deductive and Object-Oriented Databases, 5th In-
ternational Conference (DOOD’97), pages 85–102.
Broekstra, J., Kampman, A., and Harmelen, F. V. (2002).
Sesame: A generic architecture for storing and query-
ing rdf and rdf schema. In International Semantic Web
Conference, pages 54–68.
Calbimonte, J. P., Porto, F., and Keet, C. M. (2009). Func-
tional dependencies in owl abox. In Brazilian Sympo-
sium on Databases (SBBD), pages 16–30.
Calvanese, D., Giacomo, G. D., Lembo, D., Lenzerini, M.,
and Rosati, R. (2008). Path-based identification con-
straints in description logics. pages 231–241.
Chen, P. P. (1975). The enity-relationship model: Toward a
unified view of data. In VLDB, page 173.
Das, S., Chong, E. I., Eadon, G., and Srinivasan, J.
(2004). Supporting ontology-based semantic match-
ing in rdbms. In VLDB, pages 1054–1065.
Dehainsala, H., Pierra, G., and Bellatreche, L. (2007). On-
todb: An ontology-based database for data intensive
applications. In DASFAA, pages 497–508.
Jean, S., Dehainsala, H., Nguyen Xuan, D., Pierra, G.,
Bellatreche, L., and A¨ıt-Ameur, Y. (2007). Ontodb:
It is time to embed your domain ontology in your
database. In Proceedings of the 12th International
Conference on Database Systems for Advanced Ap-
plications, pages 1119–1120.
Motik, B., Horrocks, I., and Sattler, U. (2009). Bridging the
gap between owl and relational databases. Journal of
Web Semantics, 7(2):74–89.
Pan, Z. and Heflin, J. (2003). Dldb: Extending relational
databases to support semantic web queries. In The
First International Workshop on Practical and Scal-
able Semantic Systems.
Romero, O., Calvanese, D., Abell´o, A., and Rodriguez-
Muro, M. (2009). Discovering functional dependen-
cies for multidimensional design. In DOLAP, pages
1–8.
WEBIST2013-9thInternationalConferenceonWebInformationSystemsandTechnologies
164