Re: lime

Dear Armando,

see below.... it seems we are finally converging and only a few smaller 
issues are left, that is very good...

Talk to you tomorrow!

Philipp.

Am 15.07.15 um 13:07 schrieb Armando Stellato:
>
> Hi Philipp,
>
> thanks very much for the review work. This is from me only, so Manuel 
> might want to add something else.
>
> I’m more or less ok with the definition change you did on the 
> definition of avgSynonymy. Just a couple of things:
>
> 1)I think you wanted to get rid of that “synonymic LexicalEntries per 
> concept” as you felt “synonymic” as redundant with the “per concept” 
> (it is clear that they are attached to a concept, exactly because they 
> are synonymic, and vice versa). I think you mentioned something like 
> that during our call.
> I’m also in favor of being even more “plainly clear”, so OK for 
> replacing with .
>

         OK, so I repleace now the property name by the longer name: 
Average Number of Lexical Entries per Concept
>
> 2)I made in turn a very small change for a typo. I think there were a 
> missing “of” in the definition: "The average synonymy property 
> indicates the average number (+of) lexical entries that evoke a given 
> concept."
>

         OK, thanks
>
> 3)The only part which does not convince me is: “evoke a given 
> concept”. I’m not sure about the best way to render in English, but 
> with respect to the previous one, this seems to be focused on one 
> concept, as if the property is computing an average over a single 
> concept (and thus is a property of the concept itself).
> I would say more “that evoke each single concept in the concept set”. 
> Or any rephrasing in line with my improvement would be fine.
>

           Indeed, I noticed this as well, but did not find a better 
formulation. I changed it to:

     The*average number of lexical entries per concept*property 
indicates the number of lexical entries that on average evoke a concept.

        Is it better now?
>
> No major issues such as introduction of new properties from scratch 
> etc.. don’t worry ;-)
>
> the only cases might be those which were originated from an extension 
> of the domain of some properties (such as referenceDataset, or 
> lexicalEntries) to include further classes (e.g. LexicalLinkSet, or 
> ConceptualizationSet), which were considered for being split instead 
> into two properties. On these, you said you preferred to think more on 
> that after the call. If, case by case, you opted for keeping the 
> property as is with the extended domain, then no need to add any 
> further property.
>
> So, a few comments on the last email exchanges between you and Manuel 
> (take this improved scheme made by Manuel as a reference: 
> https://drive.draw.io/#G0BwvuzIAhamr9X1dwRkUyTjRFaVU , which needs 
> also to be updated on the Wiki)
>
> *Manuel*: lime:lexicalEntries
>
> - The domain of this property should be Lexicon or LexicalizationSet 
> or Conceptualization and the definition should be changed accordingly, 
> unless we want to split this property into two or more properties.
>
> *Philipp: *I changed the property definition to also include 
> ConceptualizationSet as domain. You mean ConceptualizationSet, right?
>
> *Armando*:  If I’m not taking the wrong property, it seems you wrote 
> ConceptSet, but it is ConceptualizationSet. And yes, I can reply on 
> your question: he meant ConceptualizationSet, but he just, in the 
> previous email, sent Conceptualization to use the previous name, while 
> waiting  ConceptualizationSet was waiting for approval (we agreed to 
> do so to avoid confusion). I’m not changing it on the wiki, but if you 
> agree here on ConceptSetàConceptualizationSet I can go on the wiki and 
> fix it
>

OK, changed to ConceptualizationSet
>
> *Manuel*: lime:referenceDataset - the definition should be reviewed
>
> *Philipp*: For me the definition is fine, what exactly should be reviewed?
>
> *Armando*: He was referring to the domain we extended with 
> LexicalLinkSet, which is not mentioned in the definition. We discussed 
> this in the talk, but I think you or/and John had a problem with 
> extending the domain, as you preferred this property to really mean 
> the elements which are “referenced” (i.e. use of property “:reference”).
> If fine for you to keep the property with the extended domain, then 
> the original:
>
> “The*reference dataset*property indicates the dataset that contains 
> the ontology or vocabulary elements referenced by a given lexicon and 
> thus providing the grounding vocabulary for the meaning of the lexical 
> entries in the lexicon.”
>
> Could be changed into:
>
> “The*reference dataset*property indicates the dataset that contains 
> the ontology or vocabulary elements that are either referenced by a 
> given lexicon (and thus providing the grounding vocabulary for the 
> meaning of the lexical entries in it) or linked to lexical concepts in 
> a ConceptSet by means of a LexicalLinkSet”.
>

I prefer to keep "reference dataset" only defined for LexicalizationSets 
as domain. For LexicalLinkSet we have the property conceptualDataset:

The*conceptual dataset*property relates a lexical link set or a 
conceptualization set to a corresponding concept set.

Is that not sufficient?

Further, I prefer to let "references" to refer only to the triples using 
the property "reference". It is much clearer in this way and reduced 
confusion.

I changed the definition to: The*references*property indicates the 
number of distinct references that are lexicalized in a LexicalizationSet.

And removed the LexicalLinkSet from the domain, ok?


  If I understand correctly, you would want the analogous property for 
the "evokes" property, i.e. "evocations" ;-)

Do we really need it?
>
> The part in between brackets could be dropped if you feel the 
> definition is too long.
>
> That’s all for now, I leave the word to Manuel, and will give a 
> further overview later.
>
> Cheers,
>
> Armando
>
> *From:*Philipp Cimiano [mailto:cimiano@cit-ec.uni-bielefeld.de]
> *Sent:* Wednesday, July 15, 2015 8:39 AM
> *To:* public-ontolex@w3.org
> *Subject:* lime
>
> Dear all, Armando and Manuel in particular,
>
>  I went over the lime part of the model this morning in detail and for 
> me the module is fine now.
>
> I changed the definition of avgSynonymy as follows:
>
> "The average synonymy property indicates the average number lexical 
> entries that evoke a given concept."
>
> Is that fine from you side?
>
> I need to update the ontology and examples in the git, I will do this 
> tonight. Please send me any urgent issues by tonight.
>
> In general, I really want to freeze the whole model by end of this 
> week on our Friday telco. So please do not bring any major issues, 
> introduction of new properties now.
>
> There will be the chance of revising the model afterwards. I think we 
> can be very happy with what we have provided so far.
>
> This will be version 1.0 of the lemon-ontolex model ...
>
> Greetings,
>
> Philipp.
>
>
>
> -- 
> --
> Prof. Dr. Philipp Cimiano
> AG Semantic Computing
> Exzellenzcluster für Cognitive Interaction Technology (CITEC)
> Universität Bielefeld
>   
> Tel: +49 521 106 12249
> Fax: +49 521 106 6560
> Mail:cimiano@cit-ec.uni-bielefeld.de  <mailto:cimiano@cit-ec.uni-bielefeld.de>
>   
> Office CITEC-2.307
> Universitätsstr. 21-25
> 33615 Bielefeld, NRW
> Germany

-- 
--
Prof. Dr. Philipp Cimiano
AG Semantic Computing
Exzellenzcluster für Cognitive Interaction Technology (CITEC)
Universität Bielefeld

Tel: +49 521 106 12249
Fax: +49 521 106 6560
Mail: cimiano@cit-ec.uni-bielefeld.de

Office CITEC-2.307
Universitätsstr. 21-25
33615 Bielefeld, NRW
Germany