Re: notes at contepts vs notes at terms

Hi Alistair,

> The thing is, I don't think that a class of 'non-preferred terms' in the thesaurus sense would correspond to the class of wordnet WordSenses.  The wordnet metamodel (is [1] the latest version?) has three main classes: 'Word' 'WordSense' and 'Synset'.  I think the class wn:Word (which is a super-class of wn:Collocation) is closest to the notion of a 'non-preferred term', but even that I don't think matches, because a non-preferred term is always embedded in a thesaurus, and hence represents a relationship between several entities, whereas a Word is kind of an entity in its own right ... 

I don't think I understand your response to my point. I will try to 
make myself a bit clearer, maybe that helps you to point out what I'm 
missing :-)

I refer to WN because I want to make an analogy between the different 
ways WN can be used to annotate and the different ways a SKOS 
thesaurus can be used provided there are URIs for terms.

One use of WordNet is if you want to annotate a text with a specific 
sense of "bank" occuring in a text. You need a WordSense for that, not 
Word or Synset. For another application, e.g. counting the # 
occurences of a specific Word, you would like to annotate using Words.

Analogous for SKOS thesauri:

If you would like to annotate a text with a specific non-preferred 
term, you need a URI. Annotating using the Concept is something 
different, analogous to the different meanings of annotating using a 
Synset, WordSense or Word.

BTW there is a need for a class of preferred terms also, is there a 
specific reason you focus on the non-preferred in this mail?

Yep, [1] is the most recent.

Another bonus of having URIs for Terms is that it becomes possible to 
map terms in different languages to each other, instead of just 
Concepts. This would enable mapping between a non-pref term in one 
language to a pref term in another.

> There are other alternatives to defining a class of non-preferred terms, such as e.g.

If there are no compelling reasons to avoid a class Term I would 
prefer a class solution.

Cheers,
Mark.


> [1] http://www.cs.vu.nl/~mark/wn/17-10-05/wn.rdfs

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

Received on Wednesday, 19 October 2005 16:05:09 UTC