W3C home > Mailing lists > Public > public-swd-wg@w3.org > May 2007

[SKOS] chatting on SKOS concepts and ontology classes (was Re: ISSUE-26: SimpleExtension proposal)

From: Antoine Isaac <aisaac@few.vu.nl>
Date: Thu, 31 May 2007 19:31:41 +0200
Message-ID: <465F067D.9020008@few.vu.nl>
To: Daniel Rubin <rubin@med.stanford.edu>
CC: SWD WG <public-swd-wg@w3.org>

Hi Daniel,
>>
>>>>> In the RadLex use case, for example, "terms" are entities in 
>>>>> reality, such as blood vessels. These are linked together via 
>>>>> relations such as "part-of" and "continuous-with".  It would be 
>>>>> better for these things to be called "Entity" instead of "Term".  
>>>>> I think "Entity" would be consistent with terms and things in 
>>>>> reality.
>>>> I think your "entity" proposal, I fear this is too 
>>>> ontology-oriented. SKOS exists to model knowledge organization 
>>>> schemes, as very specific intellectual constructs made of concepts 
>>>> (i.e. instances of skos:Concept), not entities in the world 
>>>> themselves. For this real world things, ontologies should be used, 
>>>> containing instances of owl:Class or rdfs:Class (notice that OWL 
>>>> features an owl:Thing that pretty much corresponds to the class of 
>>>> your "entities").
>>>
>>> I thought the goal of SKOS is to "provide a standard way to 
>>> represent knowledge organization systems." Ontologies certainly fall 
>>> under that umbrella. Some of the SKOS model is certainly very 
>>> relevant to people building ontologies.
>> Indeed. But assuming that a skos:Concept in a KOS actually denotes 
>> entities in the world in a formal way - i.e. making it a class - is 
>> an extra interpretation of your concepts.
>> For a clearer account on this things just have a look at 
>> http://www.w3.org/TR/swbp-skos-core-guide/#secmodellingrdf which 
>> tries to explain this distinction.
>
> Perhaps it would be helpful if you provided the formal definition of 
> "Concept." Is this the same thing as a "Term?" Clearly, there is a 
> difference between actual things and the way we talk about things.
OK, roughly copy-pasting from wikipedia. Not fully satisfied, but at 
least we have a glimpse of the difference of level between the two notions:
A concept is an abstract idea or a mental symbol, typically associated 
with a corresponding representation in language or symbology, that 
denotes all of the objects in a given category or class of entities, 
interactions, phenomena, or relationships between them
A "term" is a word, word pair, or word group, that is used in specific 
contexts for a specific meaning.
>
>
>> Notice however that if there is a modeling distinction, there is no 
>> exclusion. You can have skos:Concepts (my:Car rdf:type skos:Concept) 
>> that are also RDFS/OWL classes (ex:Car rdf:type rdfs:Class) so that 
>> you can create 'objects' which are classified under it (ex:danielsCar 
>> rdf:type ex:car). This would be needed by a range of applications 
>> (including RadLex) that require using SKOS features to define 
>> conceptual entities that are actually classes in ontologies.
>
> I'm confused by this statement. You say you can have skos:Concepts 
> (my:Car rdf:type skos:Concept) that are also RDFS/OWL classes (ex:Car 
> rdf:type rdfs:Class)--how can it be that a concept is also something 
> tangible such as a car?
I'm not saying that a concept can be tangible as a car, I'm saying that 
a concept can be associated to a set of things (its extension), that is, 
interpreted as a class (just as introduced in [1]).

[1] http://www.w3.org/TR/rdf-schema/#ch_classes
>
>
>> Notice that such a trick is needed because SKOS does not make any 
>> assumption on the interpretations of instances of skos:Concept. 
>> Before you say it is an RDFS/OWL class (eventually just by creating 
>> an instance of it using rdf:type), your ex:Car is not expected to 
>> denote a set of "entities" in a "real" world (i.e. a concrete one, 
>> made of all kind of things like blood vessels and cars).
>
> I'm trying to understand this in terms of the relationship between 
> skos:Concept and RDFS/OWL class. Are you saying that it is possible to 
> say that owl:Class is-a skos:Concept?
No. Here I want to say *an* owl:Class can be *a* skos:Concept (ex:Car 
rdf:type owl:Class, ex:Car rdf:type skos:Concept)
>
>
>>
>> So to sum up my understanding, SKOS is meant to describe concepts, as 
>> "something in someone's head".
>> The point is that a concept can just stay in someone's head, or at  
>> the level of a (set of) manifestation in one language. This is the 
>> level at which a skos:Concept would appear in a library, let's say. 
>> They know what a car is (skos:definition "a car is something that has 
>> four wheels, etc."), what it is semantically related to (skos:broader 
>> ex:TransportMeans) and what labels are used to name it in language 
>> (skos:prefLabel "car"@en). But their "world" do not contain any 
>> concrete car.
>
> It's fine if skos wants to stay restricted to how people "talk about 
> things", but there needs to be a formal way of relating that to 
> ontologies that contain classes representing the things themselves and 
> that also want to talk about how they are named. That's what's going 
> on in RadLex.
Ok, so I suppose that you have radlex:BloodVessel, which is an instance 
of skos:Concept. And you want to say that you will have ex:aorta as an 
instance of blood vessel, i.e. ex:aorta rdf:type radlex:BloodVessel, 
don't you?
>
>> Then, if you have another world interested in car "entities" you 
>> would have to make your car an RDFS/OWL class, that denote set of 
>> concrete car entities (ex:danielsCar rdf:type ex:Car). But here we 
>> enter the field of ontology engineering, and we don't want SKOS to be 
>> redundant with RDFS/OWL
>
> But we need to be able to give ontology people a way to say in their 
> ontology that their entities are named things, which is where skos 
> comes in. It would not be semantically correct to do this by making 
> owl Classes subclasses of skos:Concept.
*instances of* skos:Concept (rdf:type skos:Concept) was what I propose. 
*Not rdfs:subclassOf* skos:Concept, which is a different thing, and 
surely wrong from a modeling point of view, I agree.

Cheers,

Antoine
Received on Thursday, 31 May 2007 17:32:02 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:07:49 UTC