[protege-owl] complement to partition individuals

classic Classic list List threaded Threaded
12 messages Options
Reply | Threaded
Open this post in threaded view
|

[protege-owl] complement to partition individuals

Massimo Coletti
Hi,

I have a global class:

Class(Requisito complete
                        unionOf(Requisito_composito
                                Requisito_elementare))

and two subclasses:

Class(Requisito_composito complete
                                  restriction(specifica-requisito
someValuesFrom(Requisito)))

SubClassOf(Requisito_composito Requisito)

DisjointClasses(Requisito_composito Requisito_elementare)

and

Class(Requisito_elementare complete
                                   unionOf(Requisito
                                           
complementOf(Requisito_composito)))

SubClassOf(Requisito_elementare Requisito)

DisjointClasses(Requisito_elementare Requisito_composito)

The  restriction on the second class (complementOf) makes the
knowledgebase inconsistent (Pellet, RACER).

What's wrong?

Massimo


This e-mail and any attachments may contain confidential and
privileged information. If you are not the intended recipient,
please notify the sender immediately by return e-mail, delete this
e-mail and destroy any copies. Any dissemination or use of this
information by a person other than the intended recipient is
unauthorized and may be illegal.
_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl
Reply | Threaded
Open this post in threaded view
|

[protege-owl] Best Practice for multilingual ontologies

Schentz Herbert
In OWL there are a lot of different possibilities to express multilingual ontologies.
Is there any best practice for that use case ?

regards


Herbert Schentz
IT-Business Analyse
T: +43-(0)1-313 04/5308
F: +43-(0)1-313 04/3555
[hidden email]

Umweltbundesamt
Spittelauer Lände 5
1090 Wien
Österreich/Austria
http://www.umweltbundesamt.at

Die Informationen in dieser Nachricht sind vertraulich und ausschließlich für die/den AdressatIn bestimmt. Sollten
Sie diese Nachricht irrtümlich erhalten haben, benachrichtigen Sie bitte umgehend die/den SenderIn und löschen
Sie das Original. Jede andere Verwendung dieses E-Mails ist untersagt.

This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private
information. If you have received it in error, please notify the sender immediately and delete the original. Any other
use of the email by you is prohibited.
_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] Best Practice for multilingual ontologies

Olivier Dameron
On Wed, 20 Sep 2006 11:53:02 +0200, "Schentz Herbert"
<[hidden email]> wrote:

> In OWL there are a lot of different possibilities to express
> multilingual ontologies. Is there any best practice for that use
> case ?

I am not sure if it is a best practice, but I would tend to
modularize as much as possible:
- you ontology in one file (or several, with one that imports
everything): eg myOntology.owl
- each language-specific in a separate file that imports the ontology
one: myOntology-en.owl, myOntology-fr.owl, myOntology-de.owl, ... each
one imports myOntology.owl

my 0.02euros
Olivier
_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] Best Practice for multilingual ontologies

Steve Wartik
Oliver,

I agree with your idea about modularization. I am curious to know why you would have myOntology-??.owl import myOntology.owl, rather than the other way around. If myOntology.owl imports myOntology-en.owl, myOntology-fr.owl, etc. then everyone can reference a single URI, yet still have access to the ontology in their preferred language.

Your thoughts?

Steve Wartik

Olivier Dameron wrote:
On Wed, 20 Sep 2006 11:53:02 +0200, "Schentz Herbert"
[hidden email] wrote:

  
In OWL there are a lot of different possibilities to express
multilingual ontologies. Is there any best practice for that use
case ?
    

I am not sure if it is a best practice, but I would tend to
modularize as much as possible:
- you ontology in one file (or several, with one that imports
everything): eg myOntology.owl
- each language-specific in a separate file that imports the ontology
one: myOntology-en.owl, myOntology-fr.owl, myOntology-de.owl, ... each
one imports myOntology.owl

my 0.02euros
Olivier
  


_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] Best Practice for multilingual ontologies

Olivier Dameron
Quoting Steven Wartik <[hidden email]>:

> Oliver,
>
> I agree with your idea about modularization. I am curious to know why
> you would have myOntology-??.owl import myOntology.owl, rather than the
> other way around. If myOntology.owl imports myOntology-en.owl,
> myOntology-fr.owl, etc. then everyone can reference a single URI, yet
> still have access to the ontology in their preferred language.

The idea is that people may only be interested in some of the  
languages, not all, or even
no language at all. This way, they can choose whatever they like.

Second, I assume that you are not going to do all the translations in  
every language.
More likely, each language will be contributed by a different person.  
This way, thay can
all manage and share their language-specific part the way they like, without a
centralised repository. You could even have several versions of a same  
language (not sure
if that would really be useful, though, just speculating)

Eventually, I also see your point to have a bundle. Well, nothing  
prevents you from
having:
- myOntology.owl
- myOntology-en.owl, myOntology-de.owl, etc, each one importing myOntology.owl
- myOntology-multilingual.owl that imports all the language-specific ones

This way, people can choose the bare ontology, the ontology + some  
languages, or the
ontology + all languages.

PS: if at some point you want to map myOntology-de.owl with some other  
german ontology,
you obviously won't need the french part.

Cheers
Olivier

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] Best Practice for multilingual ontologies

Steve Wartik
Hmm. I think I was assuming some tool capabilities that do not currently exist, even in (gasp!) Protégé. I like the notion of a concept being independent of language -- a rose by any other name, and all that -- and want to tell my ontology editor "Use the English version" or "Use the Sanskrit version" without having to switch URIs. Then too, would you want to map only the German version of an ontology? I'd think it would be more useful to map concepts in myOntology.owl. Maybe I'm not understanding what you desire to map.

But as you point out, my approach either asks a lot of whoever is responsible for myOntology.owl, or requires some advanced capability that allows basically anyone to make certain edits to an ontology. I rather prefer my ideal world, but I may need to live in yours for the foreseeable future.

Steve Wartik

Olivier Dameron wrote:
Quoting Steven Wartik [hidden email]:

  
Oliver,

I agree with your idea about modularization. I am curious to know why
you would have myOntology-??.owl import myOntology.owl, rather than the
other way around. If myOntology.owl imports myOntology-en.owl,
myOntology-fr.owl, etc. then everyone can reference a single URI, yet
still have access to the ontology in their preferred language.
    

The idea is that people may only be interested in some of the  
languages, not all, or even
no language at all. This way, they can choose whatever they like.

Second, I assume that you are not going to do all the translations in  
every language.
More likely, each language will be contributed by a different person.  
This way, thay can
all manage and share their language-specific part the way they like, without a
centralised repository. You could even have several versions of a same  
language (not sure
if that would really be useful, though, just speculating)

Eventually, I also see your point to have a bundle. Well, nothing  
prevents you from
having:
- myOntology.owl
- myOntology-en.owl, myOntology-de.owl, etc, each one importing myOntology.owl
- myOntology-multilingual.owl that imports all the language-specific ones

This way, people can choose the bare ontology, the ontology + some  
languages, or the
ontology + all languages.

PS: if at some point you want to map myOntology-de.owl with some other  
german ontology,
you obviously won't need the french part.

Cheers
Olivier

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl


  


_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] Best Practice for multilingual ontologies

Schentz Herbert
Nachricht
Thanks Olivier, thanks Steve!
 
I learned through your discussion, that there are a lot of different things I have to take into account. Thank you for that!  I think I have to be aware of "incomplete" languages, not containing translations for all concepts and we will, even in Europe have some sort of "Master Language" (English and Latin for some concepts). 
 
Did I get you right, that I should work with separate classes and instances for each language, setting them equivalent, and not with (dc-) annotations, because you can not reason against annotations ? (I know that there are problems in graphical displays and searching of annotations)
 
regards
 
 
 
Herbert Schentz
IT-Entwicklung
IT-Development
T: +43-(0)1-313 04/5308
F: +43-(0)1-313 04/3555
[hidden email]

Umweltbundesamt
Spittelauer Lände 5
1090 Wien
Österreich/Austria
http://www.umweltbundesamt.at

-----Ursprüngliche Nachricht-----
Von: [hidden email] [mailto:[hidden email]] Im Auftrag von Steven Wartik
Gesendet: Mittwoch, 20. September 2006 23:31
An: User support for the Protege-OWL editor
Betreff: Re: [protege-owl] Best Practice for multilingual ontologies

Hmm. I think I was assuming some tool capabilities that do not currently exist, even in (gasp!) Protégé. I like the notion of a concept being independent of language -- a rose by any other name, and all that -- and want to tell my ontology editor "Use the English version" or "Use the Sanskrit version" without having to switch URIs. Then too, would you want to map only the German version of an ontology? I'd think it would be more useful to map concepts in myOntology.owl. Maybe I'm not understanding what you desire to map.

But as you point out, my approach either asks a lot of whoever is responsible for myOntology.owl, or requires some advanced capability that allows basically anyone to make certain edits to an ontology. I rather prefer my ideal world, but I may need to live in yours for the foreseeable future.

Steve Wartik

Olivier Dameron wrote:
Quoting Steven Wartik [hidden email]:

  
Oliver,

I agree with your idea about modularization. I am curious to know why
you would have myOntology-??.owl import myOntology.owl, rather than the
other way around. If myOntology.owl imports myOntology-en.owl,
myOntology-fr.owl, etc. then everyone can reference a single URI, yet
still have access to the ontology in their preferred language.
    

The idea is that people may only be interested in some of the  
languages, not all, or even
no language at all. This way, they can choose whatever they like.

Second, I assume that you are not going to do all the translations in  
every language.
More likely, each language will be contributed by a different person.  
This way, thay can
all manage and share their language-specific part the way they like, without a
centralised repository. You could even have several versions of a same  
language (not sure
if that would really be useful, though, just speculating)

Eventually, I also see your point to have a bundle. Well, nothing  
prevents you from
having:
- myOntology.owl
- myOntology-en.owl, myOntology-de.owl, etc, each one importing myOntology.owl
- myOntology-multilingual.owl that imports all the language-specific ones

This way, people can choose the bare ontology, the ontology + some  
languages, or the
ontology + all languages.

PS: if at some point you want to map myOntology-de.owl with some other  
german ontology,
you obviously won't need the french part.

Cheers
Olivier

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl


  


_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] Best Practice for multilingual ontologies

Olivier Dameron
In reply to this post by Steve Wartik
On Wed, 20 Sep 2006 17:30:40 -0400, Steven Wartik <[hidden email]>
wrote:

> I
> rather prefer my ideal world, but I may need to live in yours for the
> foreseeable future.

Hey, what's wrong with my world? ;-)
Anyway, welcome in it!
olivier
_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] Best Practice for multilingual ontologies

Olivier Dameron
In reply to this post by Schentz Herbert
On Thu, 21 Sep 2006 07:51:18 +0200, "Schentz Herbert"
<[hidden email]> wrote:

> Did I get you right, that I should work with separate classes and
> instances for each language, setting them equivalent, and not with
> (dc-) annotations, because you can not reason against annotations ?

I think Steven question was more about rdfs:label.
Anyway, you are right, DL-reasoners don't care about annotation
properties. It is not like they complain if you use some: they just
ignore them.
Olivier
_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] complement to partition individuals

Patrizia Asirelli
In reply to this post by Massimo Coletti
Dear Massimo,

  According to your definitions Requisito_elementare includes  
Requisito and thus it also includes Requisito-composito this goes  
angainst the condition that Requisito_elementare and Requisito-
composito are to be disjoint. This independently from the defintion

> Class(Requisito_elementare complete
>                                    unionOf(Requisito
>
> complementOf(Requisito_composito)))

which makes the definiton of Requisito_elementare ciclic since the  
complementOf(Requisito_composito) includes Requisito_elementare


.....

Patrizia Asirelli
ISTI Istituto di Scienza e Tecnologia dell'Informazione "Alessandro  
Faedo"
CNR
Via Moruzzi, 1 - 56124 Pisa - Italy
Tel. +39 050 315 2795; mobile: +39 348 8283127; fax: +39 050 315 2810
email: [hidden email]



On 19/set/06, at 18:08, Massimo Coletti wrote:

> Hi,
>
> I have a global class:
>
> Class(Requisito complete
>                         unionOf(Requisito_composito
>                                 Requisito_elementare))
>
> and two subclasses:
>
> Class(Requisito_composito complete
>                                   restriction(specifica-requisito
> someValuesFrom(Requisito)))
>
> SubClassOf(Requisito_composito Requisito)
>
> DisjointClasses(Requisito_composito Requisito_elementare)
>
> and
>
> Class(Requisito_elementare complete
>                                    unionOf(Requisito
>
> complementOf(Requisito_composito)))
>
> SubClassOf(Requisito_elementare Requisito)
>
> DisjointClasses(Requisito_elementare Requisito_composito)
>
> The  restriction on the second class (complementOf) makes the
> knowledgebase inconsistent (Pellet, RACER).
>
> What's wrong?
>
> Massimo
>
>
> This e-mail and any attachments may contain confidential and
> privileged information. If you are not the intended recipient,
> please notify the sender immediately by return e-mail, delete this
> e-mail and destroy any copies. Any dissemination or use of this
> information by a person other than the intended recipient is
> unauthorized and may be illegal.
> _______________________________________________
> protege-owl mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-owl

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] complement to partition individuals

Massimo Coletti
cara Patrizia,

piacere di conoscerti, e grazie per la risposta.

Patrizia Asirelli ha scritto:
> Dear Massimo,
>
>   According to your definitions Requisito_elementare includes  
> Requisito
mmh. Requisito_elementare is a subclass of Requisito - i.e. has
Requisito as necessary condition. In my understanding, this means that
all the individuals of Requisito_elementare are also individuals of
Requisito, but not the contrary. Have I misunderstood the specifications?

Saluti,

Massimo



This e-mail and any attachments may contain confidential and
privileged information. If you are not the intended recipient,
please notify the sender immediately by return e-mail, delete this
e-mail and destroy any copies. Any dissemination or use of this
information by a person other than the intended recipient is
unauthorized and may be illegal.
_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] complement to partition individuals

Patrizia Asirelli
Ciao Massimo,
piacere mio

On 21/set/06, at 16:26, Massimo Coletti wrote:

>
> mmh. Requisito_elementare is a subclass of Requisito - i.e. has
> Requisito as necessary condition. In my understanding, this means that
> all the individuals of Requisito_elementare are also individuals of
> Requisito, but not the contrary.

YES, but given that later  you also define Requisito_elementare as  
the union of Requisito and something else
this means that Requisito-elementare also includes Requisito-
composito which contradict the disjoint condition...

> Have I misunderstood the specifications?
>
> Saluti,
>
> Massimo
>
>
>
> This e-mail and any attachments may contain confidential and
> privileged information. If you are not the intended recipient,
> please notify the sender immediately by return e-mail, delete this
> e-mail and destroy any copies. Any dissemination or use of this
> information by a person other than the intended recipient is
> unauthorized and may be illegal.
> _______________________________________________
> protege-owl mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-owl

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl