W3C home > Mailing lists > Public > semantic-web@w3.org > September 2007

RE: Is the ontology structure stored seamlessly with its data?

From: Michael Schneider <schneid@fzi.de>
Date: Wed, 26 Sep 2007 19:37:18 +0200
Message-ID: <0EF30CAA69519C4CB91D01481AEA06A040A373@judith.fzi.de>
To: "Emanuele D'Arrigo" <manu3d@gmail.com>
Cc: "Semantic Web Interest Group" <semantic-web@w3.org>, "public-owl-dev-request@w3.org" <public-owl-dev@w3.org>

Hi, Emanuele!

Emanuele D'Arrigo wrote at September 26, 2007:

>Another thing that is not quite clear in my mind right now is this:
>are the sets of triplets describing the class and property hierarchies
>of an ontology normally stored seamlessly alongside the data that
>is classified and characterized by those classes and properties?

With OWL, for which an RDF mapping exists, this is technically possible
without a problem. And when you, for instance, use JENA [1], a well known
RDF framework for Java, you generally /work/ with ontology based knowledge
bases in such a way (at least in principle). 

With JENA, you typically build a view to your knowledge base in the
following way:   

  1) Create a so called "Model", which is empty at the beginning

  2) Read into this Model the RDF statements representing the axioms of your
OWL ontology

  3) Read into this Model the RDF statements of your knowledge base

A "Model" in Jena represents an RDF graph, i.e. a set of RDF triples. Now,
as long as you use a pure "Model", this only gives you a view to the
combined set of RDF triples, which come from both your OWL ontology and your
knowledge base. But if you instead use an "OntModel" (which stands for
"Ontology Model"), you get an extended view to your RDF graph: Suddenly, you
have additional API functionality to access all your OWL classes and
properties, and the (explicit) sub-relationships between them (and many
other ontology specific features). The magic behind this is that the
OntModel internally separates out all those triple subsets within the
combind RDF graph, which are RDF mappings for OWL axioms. 

So this is the situation (or at least a possible and perfectly working
situation), when you /work/ with knowledge data. This does not, however,
mean that you should also /store/ ontological and assertional data together
in the same RDF graph. I think, in most cases it will be a better strategy
to have them separately stored. Then, you can easily reuse the ontology for
different knowledge bases, and combine them /on the fly/, whenever you want
to work with them.


[1] http://jena.sourceforge.net/ (JENA project page)

Dipl.-Inform. Michael Schneider
FZI Forschungszentrum Informatik Karlsruhe
Abtl. Information Process Engineering (IPE)
Tel  : +49-721-9654-726
Fax  : +49-721-9654-727
Email: Michael.Schneider@fzi.de
Web  : http://www.fzi.de/ipe/eng/mitarbeiter.php?id=555

FZI Forschungszentrum Informatik an der Universität Karlsruhe
Haid-und-Neu-Str. 10-14, D-76131 Karlsruhe
Tel.: +49-721-9654-0, Fax: +49-721-9654-959
Stiftung des bürgerlichen Rechts
Az: 14-0563.1 Regierungspräsidium Karlsruhe
Vorstand: Rüdiger Dillmann, Michael Flor, Jivka Ovtcharova, Rudi Studer
Vorsitzender des Kuratoriums: Ministerialdirigent Günther Leßnerkraus
Received on Wednesday, 26 September 2007 17:37:45 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 07:41:59 UTC