Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[protege-discussion] Jim Likes Dogs

Thomas Schneider tschneider at informatik.uni-bremen.de
Mon Jun 25 00:13:07 PDT 2012


On 25.06.2012, at 07:17, Jim Tivy wrote:

> Using classes to express what is essentially an object property assertion on
> Jim seems wrong to me.
> 
> I counter by saying this is a shortcoming of OWL2.

Why should it be? What are you missing? At least you can't blame OWL as long as you're not clear about what exact part of "Jim likes Dogs" you're desperate to express.

If you want to make a statement about all members of the class Dog, then Timothy's two suggestions are perfectly OK and fully capture the fact that Jim likes all (or some) dog.

If you need to distinguish between liking a particular member of the class Dog and liking dogs in general, you can use a different object property and Timothy's first suggestion. This seems fine to me because then there doesn't seem to be an obvious logical connection between this new OP and the old "likes": if I like some dog, it doesn't mean that I like all dogs and vice versa. If you still see a logical connection, chances are that you can express it in OWL too.

If you insist on using object property assertions (are you sure you meant "object property assertion"?), you could introduce a new individual "dog" and use an object property assertion between "jim" and "dog". If you need to relate "dog" back to the class "Dog", you can follow Timothy's first suggestion once again, using a different OP such as "represents".

Perhaps there is a more standard way of modelling that last one, but definitely I cannot agree with your claimed "shortcoming" because I don't see what you're missing.

Cheers

Thomas

> 
>> -----Original Message-----
>> From: protege-discussion-bounces at lists.stanford.edu [mailto:protege-
>> discussion-bounces at lists.stanford.edu] On Behalf Of Timothy Redmond
>> Sent: June-24-12 9:00 PM
>> To: protege-discussion at lists.stanford.edu
>> Subject: Re: [protege-discussion] Jim Likes Dogs
>> 
>> 
>> On 06/24/2012 12:36 PM, Jim Tivy wrote:
>>> Hi Folks
>>> 
>>> Can any OWL2 knowledgeable people on this list tell me how to express:
>>> 
>>> If I have a class Dogs and an Individual Jim how do I assert a
>>> property
>>> assertion: - Jim likes Dogs.
>> 
>> The assertion still appears ambiguous to me but I can express two
> variants.
>> The first one is that Jim likes all dogs.  This can be done as follows:
>> 
>> Class: Dog
>> 
>>     SubClassOf:
>>          inverse (like) value jim
>> 
>> 
>> Now the use of the inverse property may make this a bit harder to read.
>> An alternative is to break this into two axioms and give the inverse of
> like a
>> name:
>> 
>> ObjectProperty: likedBy
>>     InverseOf:
>>         like
>> 
>> Class: Dog
>> 
>>     SubClassOf:
>>         likedBy value jim
>> 
>> 
>> Another possibility is that Jim likes some dogs.  This can be said by
> giving the
>> jim individual a type:
>> 
>> 
>> Individual: jim
>>     Types:
>>         like some Dog
>> 
>> 
>> 
>> -Timothy
>> 
>>> 
>>> Jim
>>> 
>>>> -----Original Message-----
>>>> From: protege-discussion-bounces at lists.stanford.edu [mailto:protege-
>>>> discussion-bounces at lists.stanford.edu] On Behalf Of Jim Tivy
>>>> Sent: June-22-12 5:09 PM
>>>> To: 'User support for Core Protege and the Protege-Frames editor';
> 'User
>>>> support for Core Protege and the Protege-Frames editor'
>>>> Subject: Re: [protege-discussion] Importing Time Ontology
>>>> 
>>>> Hi Brian
>>>> 
>>>> The comments about making a superclass below are a hack if you ask me.
>>>> 
>>>> I had a similar problem with OWL2.  Few people seemed to get my
>> question -
>>>> which made me think I was missing something.
>>>> However, what I figured out was:
>>>> 
>>>> - A class is not an individual.
>>>> - Therefore you have to use classes in a simple way with simple
>>> annotations
>>>> (annotations can be attached to classes but not properties and
>> annotations
>>>> are weak (eg: cannot be modelled much unlike properties))
>>>> - You should model your own hierarchy using individuals even if it is
>>> class
>>>> like.
>>>> 
>>>> There was also some idea of using the same IRI as a class as an
> individual
>>>> and making your assertions there - but this did not seem to be well
>>>> understood or supported.
>>>> So I did my own tree of individuals with isChildOf properties.
>>>> 
>>>> My fundamental question is - If Dogs is a class and Jim is an
> individual
>>> how
>>>> do you say in OWL "Jim likes Dogs"??????
>>>> 
>>>> To me, until proven otherwise all this is a shortcoming of OWL, but OWL
>> is
>>>> still useful for data modelling.  Perhaps OWL3 will solve this.
> Perhaps
>>> the
>>>> answer is to make classes an individual - that is what modern
>> programming
>>>> languages do.  In java there is a class Object and a class class.  I am
>>> not sure
>>>> how this all bears out with Descriptive Logic...
>>>> 
>>>> Jim
>>>> 
>>>> 
>>>> 
>>>>> -----Original Message-----
>>>>> From: protege-discussion-bounces at lists.stanford.edu [mailto:protege-
>>>>> discussion-bounces at lists.stanford.edu] On Behalf Of Brian Michalk
>>>>> Sent: June-22-12 8:19 AM
>>>>> To: User support for Core Protege and the Protege-Frames editor
>>>>> Subject: Re: [protege-discussion] Importing Time Ontology
>>>>> 
>>>>> Matthew, thanks for your help.
>>>>> 
>>>>> I want to restrict that value to a time domain.  How would I specify
>>>>> that
>>>> it is
>>>>> 3600 seconds?
>>>>> 
>>>>> On 6/20/2012 3:13 PM, Matthew Horridge wrote:
>>>>>> Hi Brian,
>>>>>> 
>>>>>> Just add the expression
>>>>>> 
>>>>>> "hasCookingTime value 3600"
>>>>>> 
>>>>>> as a superclass (necessary condition) of VegetarianPizza.
>>>>>> 
>>>>>> "hasCookingTime value 3600"
>>>>>> 
>>>>>> is a class expression which describes all of the individuals that
>>>>>> have a
>>>>> hasCookingTime relationship to the specific integer 3600.  Hence,
>>>>> making VegetarianPizza a subclass of this class expression makes all
>>>>> individuals
>>>> that
>>>>> are instances of VegetarianPizza also have hasCookingTime
>>>>> relationships to 3600.
>>>>>> Cheers,
>>>>>> 
>>>>>> Matthew
>>>>>> 
>>>>>> On 20 Jun 2012, at 13:01, Brian Michalk wrote:
>>>>>> 
>>>>>>> Thanks, that's the help I needed.
>>>>>>> 
>>>>>>> http://www.w3.org/2006/time is the URL I needed.
>>>>>>> 
>>>>>>> Related question:
>>>>>>> 
>>>>>>> Let's say I have a class of things, like pizzas with meat, and
>>>> vegetarian
>>>>> pizzas.  I would like to assign a value to the cooking time of a
>>>>> class.  I
>>>> don't
>>>>> want to enumerate all possible cooking times, I just want to be able
>>>>> to
>>>> set
>>>>> something like vegetarian pizza hasCookingTime 3600 seconds.
>>>>>>>   From what I can see, I can only assign something like this to
>>>> individuals.
>>>>>>> On 6/19/2012 1:31 PM, Fabio Aiub Sperotto wrote:
>>>>>>>> Hi Brian,
>>>>>>>> 
>>>>>>>> You can't find the ontologies http://www.w3.org/TR/owl-time?
>>>>>>>> 
>>>>>>>> I think there are in some sections of document, for example, look
>>>>>>>> in the final of -"OWL code for the time ontology":
>>>>>>>> http://www.w3.org/TR/owl-time/#calclock
>>>>>>>> in the link: [RDF/XML]
>>>>>>>> 
>>>>>>>> But the file will be downloaded without a extension, then you need
>>>>> renamed the file including a .owl extension. So your Protégé will
>>>> recognize
>>>>> the file.
>>>>>>>> 
>>>>>>>> 2012/6/19 Brian Michalk<michalk at awpi.com>  I'm looking to import
>>>>>>>> an ontology, which I have not done before, specifically the one for
>>>> time.
>>>>> w3.org claims to have the ontology at http://www.w3.org/TR/owl-time ,
>>>>> but
>>>> I
>>>>> can't actually locate a .owl file anywhere there.  Trying to import
>>>>> that
>>>> URL
>>>>> from Protege fails.
>>>>>>>> _______________________________________________
>>>>>>>> protege-discussion mailing list
>>>>>>>> protege-discussion at lists.stanford.edu
>>>>>>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>>>>>>> 
>>>>>>>> Instructions for unsubscribing:
>>>>>>>> http://protege.stanford.edu/doc/faq.html#01a.03
>>>>>>>> 
>>>>>>>> 
>>>>>>>> 
>>>>>>>> --
>>>>>>>> Fabio Aiub Sperotto
>>>>>>>> Mestrando em Modelagem Computacional about.me/fabiosperotto
>>>>>>>> www.twitter.com/fabio_gk
>>>>>>>> 
>>>>>>>> 
>>>>>>>> 
>>>>>>>> _______________________________________________
>>>>>>>> protege-discussion mailing list
>>>>>>>> 
>>>>>>>> protege-discussion at lists.stanford.edu
>>>>>>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>>>>>>> 
>>>>>>>> 
>>>>>>>> Instructions for unsubscribing:
>>>>>>>> http://protege.stanford.edu/doc/faq.html#01a.03
>>>>>>> _______________________________________________
>>>>>>> protege-discussion mailing list
>>>>>>> protege-discussion at lists.stanford.edu
>>>>>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>>>>>> 
>>>>>>> Instructions for unsubscribing:
>>>>> http://protege.stanford.edu/doc/faq.html#01a.03
>>>>>> _______________________________________________
>>>>>> protege-discussion mailing list
>>>>>> protege-discussion at lists.stanford.edu
>>>>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>>>>> 
>>>>>> Instructions for unsubscribing:
>>>>> http://protege.stanford.edu/doc/faq.html#01a.03
>>>>> 
>>>>> 
>>>>> _______________________________________________
>>>>> protege-discussion mailing list
>>>>> protege-discussion at lists.stanford.edu
>>>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>>>> 
>>>>> Instructions for unsubscribing:
>>>>> http://protege.stanford.edu/doc/faq.html#01a.03
>>>> _______________________________________________
>>>> protege-discussion mailing list
>>>> protege-discussion at lists.stanford.edu
>>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>>> 
>>>> Instructions for unsubscribing:
>>>> http://protege.stanford.edu/doc/faq.html#01a.03
>>> _______________________________________________
>>> protege-discussion mailing list
>>> protege-discussion at lists.stanford.edu
>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>> 
>>> Instructions for unsubscribing:
>> http://protege.stanford.edu/doc/faq.html#01a.03
>> 
>> _______________________________________________
>> protege-discussion mailing list
>> protege-discussion at lists.stanford.edu
>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>> 
>> Instructions for unsubscribing:
>> http://protege.stanford.edu/doc/faq.html#01a.03
> 
> _______________________________________________
> protege-discussion mailing list
> protege-discussion at lists.stanford.edu
> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
> 
> Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03

----------------------------------------------------------------------
Dr. Thomas Schneider
Universität Bremen, FB 03
Postfach 330440
28334 Bremen
Germany
+49 421 218-64432
http://www.informatik.uni-bremen.de/~ts/
For visits: Cartesium, Room 2.56
----------------------------------------------------------------------

Zeerust (n.)
The particular kind of datedness which afflicts things that were originally designed to look futuristic.

Douglas Adams, John Lloyd: The Deeper Meaning of Liff



More information about the protege-discussion mailing list