W3C home > Mailing lists > Public > www-rdf-interest@w3.org > December 2004

Re: OWL newbie Question

From: Benjamin Nowack <bnowack@appmosphere.com>
Date: Wed, 08 Dec 2004 12:22:39 +0000
To: Michael Engelhardt <engelh@fhtw-berlin.de>
Cc: www-rdf-interest@w3.org
Message-id: <PM-EV.20041208122239.18236.1.1D@bodkin.nuigalway.ie>

hi michael,

you may want to have a look at the SKOS[1] stuff.
SKOS allows you to model classifications which don't
really fit in the very strict constructs that OWL and
RDFS offer. IIRC, you can't use an e.g. "skos:related"
property with an owl:Class, though. but with the ACM scheme
being closer a "concept" than an "object" taxonomy, it'd
perhaps be more appropriate to use skos:broader/narrower
instead of rdfs:subClassOf anyway (i.e. it's somewhat
complicated to imagine an owl:Thing of rdf:type
"Performance and Reliability" ...)


Benjamin Nowack

Kruppstr. 100
45145 Essen, Germany


On 08.12.2004 11:09:52, Michael Engelhardt wrote:
>Hello out there,
>I am currently trying to perform my first steps in the OWL world. Since
>searching for a simple and useful example I want to build up the ACM
>classification index in OWL. My approach is to express each classification
>entry as an owl:class concept, relating them via rdfs:subClassOf. That is
>pretty simple. However the ACM classification index encodes references
>between certain entries too (e.g. B.2.3 ["Reliability, Testing, and
>Fault-Tolerance Reliability, Testing, and Fault-Tolerance"] is related to B.8
>["Performance and Reliability"]) . So my question here is how to model those
>references in OWL. I believe using the equivalentClass construct is not
>appropriate, because those classes have not same instances.
>Could anybody give me an enlightening idea please?
>thanks in advance
>Michael Engelhardt

Received on Wednesday, 8 December 2004 12:20:26 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:07:54 UTC