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] Define some interval subclass

Martin O'Connor martin.oconnor at stanford.edu
Mon Apr 30 09:20:01 PDT 2007


Another possibility is to use SWRL to define these weight 
classifications. If we have a class Weight with a hasWeightInKg 
functional datatype property we can write rules as follows:

Weight(?i) ^ hasWeightInKg(?i, ?w) ^ swrlb:lessThan(?w, 10) -> 
LessThan10Kg(?i)
Weight(?i) ^ hasWeightInKg(?i, ?w) ^ swrlb:greaterThanOrEqual(?w, 10) ^ 
swrlb:lessThanOrEqual(?w, 30) ->  WeightBetween10And30Kg(?i)
Weight(?i) ^ hasWeightInKg(?i, ?w) ^ swrlb:greaterThan(?w, 30) -> 
WeightOver30Kg(?i)

These rules will reclassify individuals of class Weight to be 
appropriate weight subclasses.

I'm not sure how natural this is though.

Martin

> Hi,
>  
> I did some research for the use case. Thomas was right, OWL can not 
> define "a interval value".
> We can define it in XSD, but I think it is hard to be maintained. The 
> xsd file is imported in owl file.
> That means, if one want to use this owl file and have to define a new 
> interval, he needs to change the xsd or import a new xsd.
>  
> I hope, there is a posibility to define it directly in owl.
>  
> Regards,
>  
> Steve
>
>  
> On 4/28/07, *Nagy Attila* <nagy.attila.1984 at gmail.com 
> <mailto:nagy.attila.1984 at gmail.com>> wrote:
>
>     If you manage to find a solution, please tell us about it. As I said
>     before, I have a quite similar problem. Thanks.
>
>     Cheers,
>
>     Attis
>
>     2007/4/27, Steve Jonathan Tirtha <sjtirtha at gmail.com
>     <mailto:sjtirtha at gmail.com>>:
>     > Ok,
>     >
>     > thanks for the advices. I'll try those approaches.
>     > I'm thinking also to use new datatype defined in xsd, but only
>     pellet
>     > supports datatype reasoning.
>     > But I'll try first, than I'll ask again.
>     >
>     > steve
>     >
>     >
>     > On 4/27/07, Thomas Russ < tar at isi.edu <mailto:tar at isi.edu>> wrote:
>     > >
>     > > On Apr 27, 2007, at 4:25 AM, Steve Jonathan Tirtha wrote:
>     > >
>     > > > Hi,
>     > > >
>     > > > How can I define subclass that represents interval of its
>     superclass.
>     > > > Example:
>     > > > I have a class "Weight" and I divide it into some subclasses:
>     > > > WeightUnder10Kg, WeightBetween10And30Kg, WeightOver30Kg.
>     > > > I can define them only as a name. But how can define that
>     the class
>     > > > WeightUnder10Kg can only have weight under 10 kg.
>     > >
>     > > As pointed out by others, this is not something that you can
>     express
>     > > using OWL.
>     > >
>     > > Some restrictions like this can be expressed either by
>     creating your
>     > > own XSD datatypes.  Or else you need to look to additional
>     languages
>     > > like SWRL, which can do some things like this.
>     > >
>     > > Or you find a modeling language more expressive than OWL.
>     > >
>     > > > NB: is there any best practice for ontology modelling. I've read
>     > > > already Ontology 101. That was pretty good, but it's still
>     not enough.
>     > >
>     > > There isn't really any complete consensus on modeling practices.
>     > >
>     > > There are some hints in the OWL tests that you can run from
>     Protege.
>     > >
>     > > There has also been methodological work on ontologies in general
>     > > (rather than OWL-specific), from U. Amsterdam (CommonKADS) and
>     from
>     > > some theoretical work (Chris Welty and Nicola Guarino --
>     OntoClean)
>     > > _______________________________________________
>     > > protege-owl mailing list
>     > > protege-owl at lists.stanford.edu
>     <mailto: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
>     > >
>     >
>     >
>     > _______________________________________________
>     > protege-owl mailing list
>     > protege-owl at lists.stanford.edu
>     <mailto:protege-owl at lists.stanford.edu>
>     > https://mailman.stanford.edu/mailman/listinfo/protege-owl
>     <https://mailman.stanford.edu/mailman/listinfo/protege-owl>
>     >
>     > Instructions for unsubscribing:
>     > http://protege.stanford.edu/doc/faq.html#01a.03
>     >
>     >
>     _______________________________________________
>     protege-owl mailing list
>     protege-owl at lists.stanford.edu <mailto:protege-owl at lists.stanford.edu>
>     https://mailman.stanford.edu/mailman/listinfo/protege-owl
>     <https://mailman.stanford.edu/mailman/listinfo/protege-owl>
>
>     Instructions for unsubscribing:
>     http://protege.stanford.edu/doc/faq.html#01a.03
>
>
>------------------------------------------------------------------------
>
>_______________________________________________
>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