usage of universal quantification

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

usage of universal quantification

Joanna.zyy
Hello,
I meet a problem about define universal quantification as learning OWL 2 and SWRL Tutorial
In the Tutorial, there is an example about reasoning that Ivan is in the Czech. How to write the axioms "hasParent only Czech SubClassOf Czech", I only can write "Czech SubClassOf hasParnt only Czech" but it falls to infer Ivan is a Czech. Would you help me?

Thanks a lot!
Joanna
Reply | Threaded
Open this post in threaded view
|

Re: usage of universal quantification

Csongor Nyulas-2
Administrator
What is Ivan and what is Czech? Is Ivan an individual? Is Czech a class?
Is Ivan an instance of Czech?
Can you share with us your ontology and/or point us to the description
of the example that you try to model?
At the moment it is hard to understand what you are trying to model, and
why should it follow from your statements that "Ivan is a Czech".

Csongor

On 04/08/2015 09:00 PM, Joanna.zyy wrote:

> Hello,
> I meet a problem about define universal quantification as learning OWL 2 and
> SWRL Tutorial
> In the Tutorial, there is an example about reasoning that Ivan is in the
> Czech. How to write the axioms "hasParent only Czech SubClassOf Czech", I
> only can write "Czech SubClassOf hasParnt only Czech" but it falls to infer
> Ivan is a Czech. Would you help me?
>
> Thanks a lot!
> Joanna
>
>
>
> --
> View this message in context: http://protege-project.136.n4.nabble.com/usage-of-universal-quantification-tp4662644.html
> Sent from the Protege User mailing list archive at Nabble.com.
> _______________________________________________
> protege-user mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-user


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

Re: usage of universal quantification

Joanna.zyy
Sorry, it is an example in the Tutorial.

Declaration(NamedIndividual (:Ivan))
ClassAssertion(:Person :Ivan)
Declaration(ObjectProperty(:hasParent))
ObjectPropertyAssertion( :hasParent :Ivan :Martin )
ObjectPropertyAssertion( :hasParent :Ivan :Lenka )

SubClassOf (:Person ObjectMaxCardinality(2 :hasParent))
Declaration(Class(:Czech))
SubClassOf(:Czech :Person)
ClassAssertion(:Czech :Lenka)
ClassAssertion(:Czech :Martin)

SubClassOf(ObjectAllValuesFrom(:hasParent :Czech) :Czech) ///
DifferentIndividuals(......)

I worked this example in Protege 4.1, but I don't know how to write the axiom above marked "///" in protege 4.1. In the tutorial, with these axioms, reasoner can infer Ivan belongs to Czech class.

Thanks
Reply | Threaded
Open this post in threaded view
|

Re: usage of universal quantification

Mike Bergman
In reply to this post by Csongor Nyulas-2
This is exactly one of the attractions for us with OWL 2 in our design
of UMBEL. Sometimes one wants to look at a node in relation to other
nodes and hierarchies, the class view. Then, in other circumstances, one
may want to describe that node as an instance with attributes, the
individual view.

The beauty of the metamodel of punning in OWL 2 is that both views are
legitimate; in fact, the given object has the same URI identity. The
OWLAPI underneath Protege will handle all of this.

Optical illusions appear as they do based on viewpoint and perspective,
even though the object has a single representation. That is how I kind
of try to grok these questions, and welcome OWL 2 as a nice sleight of
hand to still make everything decidable.

So, declare all of your Ivans as classes, set your class relationships,
and then give them an object data property. OWL 2 will pun the rest.

So, maybe the answer to the question is that Ivan is both (and still is
an instance of its class).

But, please, this design pattern is not for all, for sure. However, if
you want your ontology to act as a bridge between abstract, conceptual
views of things and concrete, attribute-laden views of things, you can
get your Ivan to swing both ways.

Mike

On 4/9/2015 12:52 AM, Csongor Nyulas wrote:

> What is Ivan and what is Czech? Is Ivan an individual? Is Czech a class?
> Is Ivan an instance of Czech?
> Can you share with us your ontology and/or point us to the description
> of the example that you try to model?
> At the moment it is hard to understand what you are trying to model, and
> why should it follow from your statements that "Ivan is a Czech".
>
> Csongor
>
> On 04/08/2015 09:00 PM, Joanna.zyy wrote:
>> Hello,
>> I meet a problem about define universal quantification as learning OWL
>> 2 and
>> SWRL Tutorial
>> In the Tutorial, there is an example about reasoning that Ivan is in the
>> Czech. How to write the axioms "hasParent only Czech SubClassOf Czech", I
>> only can write "Czech SubClassOf hasParnt only Czech" but it falls to
>> infer
>> Ivan is a Czech. Would you help me?
>>
>> Thanks a lot!
>> Joanna
>>
>>
>>
>> --
>> View this message in context:
>> http://protege-project.136.n4.nabble.com/usage-of-universal-quantification-tp4662644.html
>>
>> Sent from the Protege User mailing list archive at Nabble.com.
>> _______________________________________________
>> protege-user mailing list
>> [hidden email]
>> https://mailman.stanford.edu/mailman/listinfo/protege-user
>
>
> _______________________________________________
> protege-user mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-user
>
_______________________________________________
protege-user mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-user
Reply | Threaded
Open this post in threaded view
|

Re: usage of universal quantification

Joanna.zyy
In reply to this post by Joanna.zyy
In the protege, I can only write "hasParent only Czech" is the superclasses of "Czech" by using "superclasse" Tab in Class description. But it can only infer Martin's parent is a Czech, not Ivan (Martin's son) is a Czech. But the original purpose is person whose parents are both Czech is a Czech. I don't know if I miss something~I have been bothered by this problem for half a day~please help me~

Thank you
Reply | Threaded
Open this post in threaded view
|

Re: usage of universal quantification

Adila Alfa Krisnadhi
Hi Joanna,

You should use the General class axioms tab in class description.

--adila

On Wed, Apr 8, 2015 at 11:45 PM, Joanna.zyy <[hidden email]> wrote:
In the protege, I can only write "hasParent only Czech" is the superclasses
of "Czech" by using "superclasse" Tab in Class description. But it can only
infer Martin's parent is a Czech, not Ivan (Martin's son) is a Czech. But
the original purpose is person whose parents are both Czech is a Czech. I
don't know if I miss something~I have been bothered by this problem for half
a day~please help me~

Thank you



--
View this message in context: http://protege-project.136.n4.nabble.com/usage-of-universal-quantification-tp4662644p4662650.html
Sent from the Protege User mailing list archive at Nabble.com.
_______________________________________________
protege-user mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-user


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

Re: usage of universal quantification

Joanna.zyy
Thank you, Edila!
Yes, I find the general class axioms in Protege 5.1, and it works. why can i not find it in Protege4.1?! Isn't there such Tab in Protege 4.1? In protege 4.1, this axiom can not be expressed?

Thank you again!
Reply | Threaded
Open this post in threaded view
|

Re: usage of universal quantification

Marco Colombetti-2
In reply to this post by Joanna.zyy
Joanna,

are you aware that with the axiom

   (hasParent only Czech) SubclassOf Czech

everything that has no parents (e.g., a car, a pen, Š) turns out to be
Czech?
You may want to change your axiom to

   Person and (hasParent only Czech) SubclassOf Czech

or

   (hasParent some Thing) and (hasParent only Czech) SubclassOf Czech

Marco


On 09.04.2015 08:45 , "Joanna.zyy" <[hidden email]> wrote:

>In the protege, I can only write "hasParent only Czech" is the
>superclasses
>of "Czech" by using "superclasse" Tab in Class description. But it can
>only
>infer Martin's parent is a Czech, not Ivan (Martin's son) is a Czech. But
>the original purpose is person whose parents are both Czech is a Czech. I
>don't know if I miss something~I have been bothered by this problem for
>half
>a day~please help me~
>
>Thank you
>
>
>
>--
>View this message in context:
>http://protege-project.136.n4.nabble.com/usage-of-universal-quantification
>-tp4662644p4662650.html
>Sent from the Protege User mailing list archive at Nabble.com.
>_______________________________________________
>protege-user mailing list
>[hidden email]
>https://mailman.stanford.edu/mailman/listinfo/protege-user

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

Re: usage of universal quantification

Rafael Gonçalves
In reply to this post by Joanna.zyy
Hi Joanna,

The 'General class axioms' view has been around for a while now, and it was definitely present in Protege 4.1. If it’s not visible to you by default in Protege 4.1, then you can add it from Window > Views > Class Views > General class axioms. Such axioms, aka general concept inclusions (GCIs), have always been expressible in OWL.

Hope this helps.

Cheers,
Rafael

> On Apr 9, 2015, at 00:37, Joanna.zyy <[hidden email]> wrote:
>
> Thank you, Edila!
> Yes, I find the general class axioms in Protege 5.1, and it works. why can i
> not find it in Protege4.1?! Isn't there such Tab in Protege 4.1? In protege
> 4.1, this axiom can not be expressed?
>
> Thank you again!
>
>
>
> --
> View this message in context: http://protege-project.136.n4.nabble.com/usage-of-universal-quantification-tp4662644p4662652.html
> Sent from the Protege User mailing list archive at Nabble.com.
> _______________________________________________
> protege-user mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-user


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