W3C home > Mailing lists > Public > public-media-annotation@w3.org > March 2009

Re: about the skos mapping of vocabularies

From: Tobias Bürger <tobias.buerger@sti2.at>
Date: Tue, 31 Mar 2009 15:31:42 +0200
Message-ID: <49D21B3E.5080203@sti2.at>
To: Raphaël Troncy <Raphael.Troncy@cwi.nl>
CC: Veronique Malaise <vmalaise@few.vu.nl>, public-media-annotation@w3.org
Hi Raphael,

I think it is definitely worth looking at what the ontology alignment
people did.
Perhaps you know this already: I recently came across work on something
called PSPARQL++ which combines two SPARQL extension for data mediation
[1]. I however do not know if this already implemented now and what the
status is. But this is something which could contribute to a solution of
the mapping PA was asking for.

Definitely, asking Jerome would be a good option!

Best,

Tobias

[1] ftp://ftp.inrialpes.fr/pub/exmo/publications/euzenat2008g.pdf

Raphaël Troncy schrieb:
> Hi all,
>
>> One  question, form a non-specialist, to answer PA's comment at the
>> last teleconf: he said, if I understood correctly, that it might be
>> very hard to have an API using mappings defined in SKOS. The
>> Cliopatria API [1] (also described more extensively in papers like
>> [2]) seems, again to my understanding, to be using vocabularies and
>> vocabularies mappings (in SKOS) to access and display the metadata of
>> objects from different collections, described with their own
>> description schema. Would Cliopatria be an interesting API to have a
>> closer look in out own case?
>
> The statement about Cliopatria is true.
> The ESW wiki maintains a list of tools implementing SKOS, see
> http://esw.w3.org/topic/SkosDev/ToolShed
>
> Yes, the SKOS/OWL mappings do not necessarily say everything about
> mappings. The famous example is:
>   . Vocabulary A defines: firstName and lastName properties
>   . Vocabulary B defines: name property
> Ideally, we would like to state:
>   b:name = a:firstName + a:lastName where the semantics of the '+' is
> here a string concatenation
> We could already say:
>   a:firstName rdfs:subPropertyOf b:name
>   a:lastName rdfs:subPropertyOf b:name
> or an equivalent in SKOS.
>
> The problem mentioned by PA is somehow similar: a conversion between
> units. There are plenty of algebra that provides the constructs for
> expressing such statements. XPATH is built on them.
> It is worth to investigate what the ontology alignment community has
> produced since this problem has been already researched. Actually,
> someone might have already propose a way of representing exactly that
> in OWL. Pierre Antoine, is this a question you could ask to Jérôme?
> Cheers.
>
>   Raphaël
>

-- 
_________________________________________________
Dipl.-Inf. Univ. Tobias Bürger

STI Innsbruck
University of Innsbruck, Austria
http://www.sti-innsbruck.at/

tobias.buerger@sti2.at
__________________________________________________
Received on Tuesday, 31 March 2009 13:29:05 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 31 March 2009 13:29:06 GMT