Re: Rappresentation of special compound term

Hi Piero,

You are right that there is no standard way in SKOS to represent 
compound terms and USE references to/from compounds.

In a recent paper I wrote (with Veronique Malaise, Alistair Miles and 
Guus Schreiber - will post it on the list soon) we solve a similar 
case for MeSH by extending SKOS in the following way:

- mesh:CompoundConcept rdfs:subClassOf skos:Concept
- mesh:Qualifier rdfs:subClassOf skos:Concept
- mesh:qualifier rdfs:domain mesh:Qualifier, rdfs:range skos:Concept
- mesh:main rdfs:domain mesh:CompoundConcept, rdfs:range skos:Concept
- mesh:preferredCombination rdfs:domain mesh:CompoundConcept, 
rdfs:range skos:Concept

This allows us to represent CompoundConcepts (in the MeSH case 
combinations of so-called Qualifiers and Descriptors) and refer them 
to other Concepts or CompoundConcepts.

The non-standard part is the preferredCombination property because 
there is no valid superproperty for it in SKOS. Hence this appraoch 
does the job for local purposes, but other SKOS software will not be 
aware of the reference.

If I'm correct this only partly solves your problem as the "rail 
safety" cannot be represented as a skos:Concept itself...

Hope this helps,
Mark.

PCavaleri@liuc.it wrote:
> I'm trying to convert our thesaurus in SKOS format, but I'm not able to
> rappresent the splitting of compound not preferred terms. Sometimes it may
> be usefull because users search using them.
> Using an English example, I want to allow the user to find the compound
> term rail safety, but redirecting them to use the two term railways and
> safety.
> 
> In a paper thesaurus I could write:
> rail safety (not accepted)
> use railways + safety
> 
> In SKOS the non accepted terms haven't an autonomous entry, but I can't
> indicate that rail safety is an <altlabel> for railways or for safety.
> How do you resolve problems like this?
> 
> Thank you in advance,
> 
> Piero Cavaleri
> Biblioteca Rostoni
> Università Cattaneo
> C.so Matteotti, 22
> 21053 Castellanza (VA)
> Italy
> 
> e-mail: pcavaleri@liuc.it
> 
> 
> 

-- 
  Mark F.J. van Assem - Vrije Universiteit Amsterdam
        markREMOVE@cs.vu.nl - http://www.cs.vu.nl/~mark

Received on Thursday, 23 March 2006 10:14:48 UTC