Re: WordNet modelling in Lemon and SKOS

On Sun, Apr 14, 2013 at 12:02 PM, Elena Montiel Ponsoda <
elemontiel@gmail.com> wrote:

> Dear John,
>
> thanks for this. I have some doubts/questions regarding this example that
> may be due to the fact that I have missed some telcos.
> If that is the case, my apologies.
>
> Are there no mechanisims in lemon to say that catted is the past
> participle form of "to cat", and cating is the gerund form. I think that
> the use of canonical form vs. other form is well suited for nouns, but
> somehow strange for verbs, what do you think?
>
Yes, there are but WordNet itself does not distinguish between these forms.
I don't really see why it is different for verbs as opposed to nouns
though.

>
> Regarding the use of skos, if I understand it correctly, it is intended to
> represent the lexical relations in Wordnet, right? For example, the
> hypernonymy-hyponymy relation would correspond to the skos:broader or
> skos:narrower relations. Is that so? In this way, skos would be used for
> the conceptual part, I mean, for transforming the lexical relations into
> ontological relations?
>
Yeah, WordNet synsets we generally agree do not correspond to a concept as
strong as a class in OWL, but SKOS's concept hierachy is basically
conceptually identical. Hence, we are using lemon not so much as a
lexicon-ontology but a lexicon-KOS.

>
> Howerver, for the definition or gloss, shouldn't we use lemon:definition.
> In this way, we would be using SKOS for the conceptual part and lemon for
> the lexical-linguistic part, would that be the idea? As I said, maybe I am
> missing something.
>
Perhaps, but lemon:definition is a sense level property and the synset
glosses are on the synset level, hence it makes more sense to use the SKOS
property.

>
> Finally, wouldn't it be nice to represent in the lemon:sense class that
> "to cat" is the colloquial use for vomiting? I think that since we are able
> to represent such contextual restrictions, we would really make a
> difference by doing it, don't you think so?
>
WordNet does have a property for this and it is exported properly, however
it seems to be missing from this synset (hence this example).

Regards,
John

>
> Just my two cents!
> Best,
> Elena.
>
> El 12/04/2013 16:10, John McCrae escribió:
>
>  Hi all,
>>
>> Here is the proposed modelling of WordNet as lemon and SKOS (using
>> skos:Concept for synsets)
>>
>> http://www.w3.org/community/**ontolex/wiki/Specification_of_**
>> Requirements/Linked_Data#**Example:_WordNet_as_lemon-SKOS<http://www.w3.org/community/ontolex/wiki/Specification_of_Requirements/Linked_Data#Example:_WordNet_as_lemon-SKOS>
>>
>> Any comments?
>>
>> Regards,
>> John
>>
>
>
> --
> Elena Montiel-Ponsoda
> Ontology Engineering Group (OEG)
> Departamento de Inteligencia Artificial
> Facultad de Informática
> Campus de Montegancedo s/n
> Boadilla del Monte-28660 Madrid, España
> www.oeg-upm.net
> Tel. (+34) 91 336 36 70
> Fax  (+34) 91 352 48 19
>
>