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] Query: Subclasses or Sibling

Ronald Cornet r.cornet at amc.uva.nl
Tue Apr 3 03:12:36 PDT 2007


William,

Regarding method 1:
I'm not fond of "poisoning" an ontology with all kinds of probe-classes or instances-represented-as-classes.

Moreover, still the question remains: is it possible in Protégé to express closure at an instance level (for whatever purpose...)

Regarding method 2:
I'll have to look into this, I don't know much about SWRL.
Thanks for this suggestion!

Regards, Ronald


###############################################################
Ronald Cornet, PhD                    email: R.Cornet at amc.uva.nl
dept. of Medical Informatics          phone: +31 (0)20 566 5188
Academic Medical Center, Room J1B-115 fax:   +31 (0)20 691 9840
P.O.Box 22700                         www: http://kik.amc.uva.nl/home/rcornet/
1100 DE  Amsterdam
The Netherlands                       'The truth is out there' 

> -----Original Message-----
> From: william fitzgerald [mailto:wfitzgerald at tssg.org] 
> Sent: Tuesday, April 03, 2007 12:00
> To: R.Cornet at amc.uva.nl; User support for the Protege-OWL editor
> Subject: Re: [protege-owl] Query: Subclasses or Sibling
> 
> Ronald,
> 
> Two possible solutions but we need the experts here on this one:
> 
> Method1:
> You can still kind of answer your question using DL reasoning 
> via CLASSES as opposed to instances.
> 
> So you create an individual of the class Pizza, under 
> subsumption that individual should be re-organized in the 
> hierarchy however lets suppose you defined a COMPLETE class 
> in Pizza called:
> 
> MenuToppingChoice \equiv
> \exist hasTopping Cheese \sqcap
> \exist hasTopping Tomato \sqcap
> 
> Note no closure.
> 
> given the RonaldPizza definition (previouse email) the 
> reasoner will now classify it under MenuToppingChoice also. 
> hence any pizza instance that is a RonaldPizza instance is an 
> instance of your query or probe class MenuToppingChoice and 
> hence you found your pizza!
> 
> Method 2:
> 
> I think dealing with individuals is more a rules engine thing 
> than a DL reasoner area.
> 
> So you could use SWRL to reason/infer over individuals. From 
> what I can tell SWRL will answer questions like: Ronald has a 
> brother called Bob, Ronald has a daughter called Alice. is 
> Bob an uncle? SWRL, i believe can answer those questions.
> 
> Protege has a SWRL plugin tab.
> 
> I hope the experts in ontologies can provide you with a 
> definitive answer here.
> 
> regards,
> Will.
> 
> 
> Ronald Cornet wrote:
> >> Agree?
> >>
> >
> > Yes and no.
> > I agree it works the way you do it.
> > But I like to do it this way:
> > I instantiate "Pizza" (NOT RonaldPizza) and put cheese and 
> tomato on it.
> > I want to ask the reasoner: what is the 'name' of this 
> pizza, and get the answer "RonaldPizza".
> > As long as I don't make explicit that this instantiated 
> Pizza only contains cheese and tomato, it will not be 
> recognized as a RonaldPizza.
> >
> > Agree?
> > Anyone knows a solution?
> >
> > Ronald
> >
> >
> >
> >> Hi Ronald,
> >>
> >> I would have thought that you wouldn't need to do this.
> >>
> >> That is, you defined a class of anonymous individuals that have 
> >> relationships to other individuals of other classes.
> >> hence any instance of a class then must comply with those class 
> >> restrictions in order to be a member.
> >>
> >> so a Class RonaldPizza that has relationship restrictions 
> as follows:
> >> \exist hasTopping Cheese \sqcap
> >> \exist hasTopping Tomato \sqcap
> >> \forall hasTopping(Cheese \sqcup Tomato)
> >>
> >> So RonaldPizza is restricted to having Cheese and Tomato and only 
> >> those toppings. if an instance or individual is a member of this 
> >> class then it must comply with those restrictions.
> >> And it complies with different levels of compliance based on 
> >> RoanldPizza been primitive or complete.
> >>
> >> So an instance ronpizza1 of RonaldPizza when placed in your mouth 
> >> must only have a taste of tomato and cheese and nothing else.
> >>
> >> Hence no need to close off instances per say.
> >>
> >>
> >>
> >> Ronald Cornet wrote:
> >>
> >>>> I wonder if we have any Pizza developers ie. the Manchester 
> >>>> University gurus out there that cans shed some light on
> >>>>
> >> our queries.
> >>
> >>> I hope so!
> >>>
> >>>
> >>>
> >>>> I am not sure if I fully understand item 2 of your
> >>>>
> >> request, but when
> >>
> >>>> I build my ontology in Protege, weather a class is primitive or 
> >>>> complete I use the closure axiom (\forall) across the
> >>>>
> >> properties that
> >>
> >>>> I want to restrict to certain classes. In the Classes GUI
> >>>>
> >> on the left
> >>
> >>>> are a hierarchy of classes and on the right you have property 
> >>>> restrictions.
> >>>> right click on a \exists property for example \exists
> >>>>
> >> hasTopping XYX
> >>
> >>>> and then scroll to option for axiom closure.
> >>>>
> >>>> Has this answered your question?
> >>>>
> >>>>
> >>> No, I understand how to do this with classes, as you 
> describe above.
> >>> I do not understand how to do this with instances.
> >>>
> >>> So I instantiate a pizza, put some ingredients on it, and
> >>>
> >> then want to express there aren't any other ingredients.
> >>
> >>> Ronald
> >>>
> >>>
> >>>
> >> --
> >> William M. Fitzgerald,
> >> PhD Student,
> >> Telecommunications Software & Systems Group, Waterford 
> Institute of 
> >> Technology, Cork Rd.
> >> Waterford.
> >> Office Ph: +353 51 302937
> >> Mobile Ph: +353 87 9527083
> >>
> >> _______________________________________________
> >> 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
> >>
> >>
> >>
> >
> > _______________________________________________
> > 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
> >
> >
> 
> --
> William M. Fitzgerald,
> PhD Student,
> Telecommunications Software & Systems Group, Waterford 
> Institute of Technology, Cork Rd.
> Waterford.
> Office Ph: +353 51 302937
> Mobile Ph: +353 87 9527083
> 
> 
> 




More information about the protege-owl mailing list