W3C home > Mailing lists > Public > public-ontolex@w3.org > July 2013

R: telco tomorrow, 15:00 CET, random talk

From: Armando Stellato <stellato@info.uniroma2.it>
Date: Sun, 14 Jul 2013 02:03:07 +0200
Message-ID: <51e1eac3.c2ce0e0a.0963.0d23@mx.google.com>
To: Aldo Gangemi <aldo.gangemi@cnr.it>
Cc: "'QUATTRI, Francesca [11901993r]'" <francesca.quattri@connect.polyu.hk>, 'Philipp Cimiano' <cimiano@cit-ec.uni-bielefeld.de>, 'John McCrae' <jmccrae@cit-ec.uni-bielefeld.de>, "public-ontolex@w3.org" <public-ontolex@w3.org>
Hi Aldo,

Thanks a lot for the clarification! Well, to me, it really makes sense that ontolex:reference can be applicable to owl:Ontology in the same way as to any ontology entity. I would really avoid any complexification for this. After all, the (be it formally specified or not) range of obtolex:reference already includes classes, properties, skos concepts...why not ontologies too? :)

----- Messaggio originale -----
Da: "Aldo Gangemi" <aldo.gangemi@cnr.it>
Inviato: ‎13/‎07/‎2013 23.40
A: "Armando Stellato" <stellato@info.uniroma2.it>
Cc: "Aldo Gangemi" <aldo.gangemi@cnr.it>; "'QUATTRI, Francesca [11901993r]'" <francesca.quattri@connect.polyu.hk>; "'Philipp Cimiano'" <cimiano@cit-ec.uni-bielefeld.de>; "'John McCrae'" <jmccrae@cit-ec.uni-bielefeld.de>; "public-ontolex@w3.org" <public-ontolex@w3.org>
Oggetto: Re: telco tomorrow, 15:00 CET, random talk

I am thinking about the relation between a sense and an entire (typically small) ontology that formalizes a gloss. In that use case, no specific ontology element is the ontolex:reference of the sense, but the whole ontology is (remember that owl:Ontology is an OWL primitive). 
If we accept that ontolex:reference can also hold for ontologies (not only for ontology entities), fine, otherwise we have to think about another relation.
Aldo

On Jul 12, 2013, at 11:34:36 PM , "Armando Stellato" <stellato@info.uniroma2.it> wrote:

> Hi Aldo. I was thinking about that too (in terms of "is it the case to think
> of some axiom for bringing a lexicon glosses automatically to the
> ontology?"), though actually I'm not sure if I understood the exact property
> you are speaking about.
> Currently, we already have a property for linking senses directly to
> ontology entities (ontolex:reference).
> So maybe you were considering having a direct link from glosses of the
> senses to the ontology elements ontolex:referenced by these senses? ...and
> in case, having it automatically inferred through an axiom?
> 
> 
>> -----Original Message-----
>> From: Aldo Gangemi [mailto:aldo.gangemi@cnr.it]
>> Sent: Friday, July 12, 2013 8:27 PM
>> To: Armando Stellato
>> Cc: Aldo Gangemi; 'QUATTRI, Francesca [11901993r]'; 'Philipp Cimiano';
> 'John
>> McCrae'; public-ontolex@w3.org
>> Subject: Re: telco tomorrow, 15:00 CET, random talk
>> 
>> Hi all, sorry for today.
>> This is interesting; actually sometimes I pointed out that glosses are
> actually
>> senses, though expressed verbosely and not with clear cut identifiers.
>> 
>> Anyway, when representing the structure of a traditional dictionary, we
> need to
>> create identifiers for different senses of a lemma, and at that point, the
> gloss
>> can be attached to sense identifiers through the gloss datatype property.
>> 
>> Once we have that, gloss analysis can be conducted, and a formal
> definition
>> can be extracted that makes it explicit the ontology attached to the
> sense.
>> 
>> In such extensions (e.g. Mihalcea's or Hovy's gloss formalizations, or
> Tìpalo-
>> FRED RDFization of Wikipedia definitions), a special relation could link
> the
>> sense (with its gloss) to the ontology formalizing it. Should such a
> relation
>> should be considered in OntoLex, or left to possible extensions?
>> 
>> Ciao
>> Aldo
>> 
>> On Jul 12, 2013, at 4:51:21 PM , "Armando Stellato"
>> <stellato@info.uniroma2.it> wrote:
>> 
>>> Hi Francesca,
>>> 
>>> in replying to Guido - who was advocating the possibility of linking
>>> glosses to different entries (LexicalSenses, or LexicalConcepts) - I
>>> said: "you are right Guido, as there are lexical resources which have
>>> no notion of LexicalConcept, think about Dictionaries (either
>>> bilingual or monolingual) which just have entries, and sense-separated
>>> descriptions, which may contain morphological variations, synonyms
>>> (translations for bilingual dictionaries), glosses etc..". Thus in
>>> Dictionaries, there are just lexical entries, and their descriptions
>>> which are sense-separated, but there is no gluing object for senses.
>>> There is even no guarantee that two senses of two lexical entries,
>>> which ideally collapse into a same meaning (LexicalConcept), have the
>>> same gloss, because these are handled separately in the descriptions
>>> of the two lexical entries (though, hopefully, the two glosses will
>>> provide very similar descriptions :-) ). For these resources, IF we
>>> want to represent them, there is no choice but allowing for glosses to
> be
>> attached to LexicalSenses.
>>> 
>>> My suggestion was to use the metadata, to understand which kind of
>>> lexical resource we are dealing with, and thus know in advance where
>>> the glosses (if
>>> any) are attached to.
>>> 
>>> Best,
>>> 
>>> Armando
>>> 
>>> 
>>> 
>>> 
>>>> -----Original Message-----
>>>> From: QUATTRI, Francesca [11901993r]
>>>> [mailto:francesca.quattri@connect.polyu.hk]
>>>> Sent: Friday, July 12, 2013 3:47 PM
>>>> To: Philipp Cimiano
>>>> Cc: John McCrae; Aldo Gangemi; public-ontolex@w3.org
>>>> Subject: RE: telco tomorrow, 15:00 CET, random talk
>>>> 
>>>> To keep up with tonight's discussion:
>>>> 
>>>> I agree with Guido's note on different meanings for a same lexical
> entry:
>>> This
>>>> occurs in one language and of course particularly across languages: I
>>>> have
>>> no
>>>> practical reference for Guido's example "dog-Hund", but for instance
>>>> the Chinese entry of 'dog' should include, apart from "domesticated
>>>> animal", "edible animal", since dogs are commonly eaten.
>>>> 
>>>> Citing Armando: "Sometimes senses are not factorized on the WN
>>>> glosses" -
>>> if I
>>>> got it right, can you give us an example?
>>>> 
>>>> F.
>>>> ________________________________________
>>>> From: QUATTRI, Francesca [11901993r]
>>>> Sent: Friday, July 12, 2013 9:29 PM
>>>> To: Philipp Cimiano
>>>> Cc: John McCrae; Aldo Gangemi; public-ontolex@w3.org
>>>> Subject: RE: telco tomorrow, 15:00 CET, random talk
>>>> 
>>>> Hi and sorry for the bad Skype connection.
>>>> Here it comes again.
>>>> F.
>>>> ________________________________________
>>>> From: QUATTRI, Francesca [11901993r]
>>>> Sent: Friday, July 12, 2013 9:06 PM
>>>> To: Philipp Cimiano
>>>> Cc: John McCrae; Aldo Gangemi
>>>> Subject: RE: telco tomorrow, 15:00 CET, random talk
>>>> 
>>>> Was playing around with the model. Thanks Philip for the example.
>>>> Take the following as a random talk about the many implications or
>>> extensions
>>>> that can derive from it.
>>>> 
>>>> Let's assume sb is not looking for the French puddle, but starts from
>>> 'dog' as
>>>> point of discussion and tries to derive analogies across languages
>>>> from
>>> its
>>>> inflections.
>>>> 
>>>> Let's assume we look for a mapping of 'dogged' (stubbornly
>>>> relentless,
>>>> persistent):
>>>> we find similar concepts in other languages (perse2ve2rance,
>>>> obstination
>>> -fr;
>>>> perseverante, ostinato -it; hartnaeckig, verbissen- de > interestingly:
>>> verbissen
>>>> from Biss - bite; hartnaeckig / probably from Nacken - back, lit.
>>>> hard
>>> back >
>>>> similar expression in It: "avere le spalle forti" /lit. to have
>>>> strong
>>> shoulders)
>>>> 
>>>> Let's go for "to be dogged" (e.g. to be dogged by an illness) We have
>>>> the concept of 'persecution' in at least four languages:
>>>> *ser maltratado por/ser castigado por/ser perseguidado por (Sp)
>>>> *verfolgt
>>> von
>>>> (Ge) (to be persecuted) *zhe2mo (persecution, torment)(Ch); wei3sui2
> (lit.
>>> "tail
>>>> behind") versus the normal gou3 ("dog") *perseguitato, maltrattato
>>>> (It)
>>>> 
>>>> Let's look for a collocation with the word, e.g. "to dog around":
>>>> Here we
>>> have at
>>>> least two meanings.
>>>> 1.to work hard 2. to cheat on sb (dogging, slang: a woman picking up
>>>> men
>>> at
>>>> random)
>>>> 
>>>> if we go for adj. plus word (e.g. top dog), we also get another new
>>> meaning (in
>>>> this case: the leader or chief of a group). Interestingly, in German
>>>> we
>>> don't
>>>> have the dog but the deer or stag to denote the concept (Platzhirsch).
>>>> 
>>>> 
>>>> ________________________________________
>>>> From: Philipp Cimiano [cimiano@cit-ec.uni-bielefeld.de]
>>>> Sent: Friday, July 12, 2013 3:36 AM
>>>> To: public-ontolex@w3.org
>>>> Subject: Re: telco tomorrow, 15:00 CET
>>>> 
>>>> Sorry, I forgot the diagram with the example.
>>>> 
>>>> Apologies,
>>>> 
>>>> Philipp.
>>>> 
>>>> Am 11.07.13 21:33, schrieb Philipp Cimiano:
>>>>> Dear all,
>>>>> 
>>>>> John sent around a link to the current version of the model early
>>>>> this
>>>>> week:
>>>>> 
>>>>> http://www.w3.org/community/ontolex/wiki/OntoLex_Core_Model
>>>>> 
>>>>> I attach an illustrative example to this mail that shows how the
>>>>> model would put into action. Hope this helps.
>>>>> 
>>>>> Tomorrow we will have our regular telco at 15:00 (CET).
>>>>> 
>>>>> I will ask everyone on the telco to raise final issues with the model.
>>>>> If there are no issues, we will then start the voting procedure
>>>>> involving the whole list.
>>>>> 
>>>>> Best regards,
>>>>> 
>>>>> Philipp.
>>>>> 
>>>> 
>>>> 
>>>> --
>>>> Prof. Dr. Philipp Cimiano
>>>> Semantic Computing Group
>>>> Excellence Cluster - Cognitive Interaction Technology (CITEC)
>>>> University
>>> of
>>>> Bielefeld
>>>> 
>>>> Phone: +49 521 106 12249
>>>> Fax: +49 521 106 12412
>>>> Mail: cimiano@cit-ec.uni-bielefeld.de
>>>> 
>>>> Room H-127
>>>> Morgenbreede 39
>>>> 33615 Bielefeld
>>>> 
>>>> 
>>> ----------------------------------------------------------------------
>>> ------
>>> ----------------------
>>>> 
>>> ----------------------------------------------------------------------
>>> ------
>>> ---
>>>>       Disclaimer:
>>>> 
>>>> This message (including any attachments) contains confidential
>>>> information intended for a specific individual and purpose. If you
>>>> are not the
>>> intended
>>>> recipient, you should delete this message and notify the sender and
>>>> the University immediately. Any disclosure, copying, or distribution
>>>> of this
>>> message,
>>>> or the taking of any action based on it, is strictly prohibited and
>>>> may be unlawful.
>>>> 
>>>> The University specifically denies any responsibility for the
>>>> accuracy or
>>> quality
>>>> of information obtained through University E-mail Facilities. Any
>>>> views
>>> and
>>>> opinions expressed are only those of the author(s) and do not
>>>> necessarily represent those of the University and the University
>>>> accepts no liability whatsoever for any losses or damages incurred or
>>>> caused to any party as a result of the use of such information.
>>>> 
>>> 
>>> 
> 
Received on Sunday, 14 July 2013 00:03:47 UTC

This archive was generated by hypermail 2.3.1 : Monday, 23 October 2017 10:57:30 UTC