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-owl] Defining same individuals and a question in defined classes

Timothy Redmond tredmond at stanford.edu
Wed Apr 25 16:02:21 PDT 2007


Hello,

I was wondering if you could help us by sending the ontology?

> Question 1 : How can I define soccer and football to be equivalent
> individuals ? If I use sameAs , Pellet generates errors and ignores  
> them.

owl:sameAs is the right thing to use.  You can assert these in  
Protege using the Individuals tab.  Click on the fifth icon from the  
left in the individual editor and enter the triple.

Now the error message

 >  SameAs construct for individuals is not supported

is a little misleading.  It turns out that this is related to a  
limitation of dig and not a limitation of Pellet.  I checked this by  
making football the same as soccer and then creating the probe class

     { football } and not { soccer }

I ran this through pellet directly using swoop and indeed this class  
was inconsistent.  In protege 4 this is handled better (much like  
swoop is doing now) because the inferences are not passed through dig.

> Question 2: After using Pellet to infer individuals, all the 3  
> individuals are
> referred under InternationalSports ; how can I make Ind1football to  
> also be
> referred under SportsKeyWord ? how to do the same for Ind2France,  
> to make it
> also be referred under CountryKeyWord as well ?

I think that this is an open world vs. closed world question.  If I  
understand right I think that you are trying to infer that

     Ind2France \in CountryKeyWord

from the following assertions

     Ind2France keyValue france

     france : Country

     CountryKeyWord == (KeyWord and (keyValue all Country)

The problem is that you cannot draw this conclusion.  Although all  
the known keyValues of Ind2France are countries it remains possible  
that there arre some unknown keyValues for Ind2France that are not  
Countries.  Therefore it is impossible to conclude on this evidence that

     Ind2France :  (keyValue all Country).

But if you made keyValue a functional property (removing the  
possibility of alternative values) then you could draw this conclusion.

-Timothy



On Apr 23, 2007, at 10:47 PM, Anita wrote:

> Hello everyone
>
> I am working on a small search utility in Protege 3.2.1. Below are  
> the list of
> my classes and properties:
>
> Class Sports - individuals: soccer, football and basketball
> Class Country - individuals: England, Brazil ,U.S.A., France
>
> Class KeyWord - individuals: Ind1football(keyValue football),  
> Ind2France
> (keyValue France), Ind3Englandfootball (keyValue England and  
> football - 2
> values)
>
>
> --------------------------------------------------------
> under KeyWord class I have 3 defined classes :
>
> SportsKeyWord : keyValue only Sports (plus KeyWord - both in  
> necessary and
> sufficient  section)
>
> CountryKeyWord : keyValue only country (plus KeyWord - both in  
> necessary and
> sufficient  section)
>
> InternationalSports : keyValue some (Sports OR Country) (plus  
> KeyWord - both
> in necessary and sufficient  section)
>
> -------------------------------------------------------
> I also the property:
> keyValue : Domain KeyWord , Range Sports + Country.
>
> ------------------------------------------------------
> Question 1 : How can I define soccer and football to be equivalent
> individuals ? If I use sameAs , Pellet generates errors and ignores  
> them.
>
> Question 2: After using Pellet to infer individuals, all the 3  
> individuals are
> referred under InternationalSports ; how can I make Ind1football to  
> also be
> referred under SportsKeyWord ? how to do the same for Ind2France,  
> to make it
> also be referred under CountryKeyWord as well ?
>
>
> I'd be deeply grateful for your help.
>
> Thank you
> Anita
>
>
>
>
> _______________________________________________
> protege-owl mailing list
> protege-owl at lists.stanford.edu
> https://mailman.stanford.edu/mailman/listinfo/protege-owl
>
> Instructions for unsubscribing: http://protege.stanford.edu/doc/ 
> faq.html#01a.03




More information about the protege-owl mailing list