Referencing the same individual in superclass expressions

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

Referencing the same individual in superclass expressions

Christian Boelling

The following expressions shall be used to define a superclass for a named class C1:

 

p1 some (C2 and (p2 some C4))

p1 some (C3 and (p2 some C4))

 

with p1, p2 being properties, C2-C4 being other classes. What needs to be added is the assertion that in both cases the same individual from C4 shall be referenced via p2. How can this be done?

 

Christian

 

 


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

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03
Reply | Threaded
Open this post in threaded view
|

Re: Referencing the same individual in superclass expressions

Martin Kuba
Dne 21.11.2011 11:34, Christian Bölling napsal(a):

> The following expressions shall be used to define a superclass for a
> named class C1:
>
> p1 some (C2 and (p2 some C4))
>
> p1 some (C3 and (p2 some C4))
>
> with p1, p2 being properties, C2-C4 being other classes. What needs to
> be added is the assertion that in both cases the same individual from C4
> shall be referenced via p2. How can this be done?

Hi Christian,

if the individual is a concrete named individual, say I1, then

p1 some (C2 and (p2 value I1))
p1 some (C3 and (p2 value I1))

However if it should be any individual,then that can't be expressed in
OWL, as that is not a tree-like structure. You need to use a SWRL for that:

p1(?a,?b),C2(?b),p2(?b,?c),C4(?c),p1(?a,?d),C3(?d),p2(?d,?c) -> C1(?a)

Best regards

Martin
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Supercomputing Center Brno             Martin Kuba
Institute of Computer Science    email: [hidden email]
Masaryk University             http://www.ics.muni.cz/~makub/
Botanicka 68a, 60200 Brno, CZ     mobil: +420-603-533775
--------------------------------------------------------------

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

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03
Reply | Threaded
Open this post in threaded view
|

Re: Referencing the same individual in superclass expressions

Matthew Horridge-2
Administrator
In reply to this post by Christian Boelling
Hi Christian,

If you make p1 and p2 functional this would work.  This might be too restrictive though.  Are C2, C3 arbitrary classes?

Cheers,

Matthew 

On 21 Nov 2011, at 02:34, Christian Bölling <[hidden email]> wrote:

The following expressions shall be used to define a superclass for a named class C1:

 

p1 some (C2 and (p2 some C4))

p1 some (C3 and (p2 some C4))

 

with p1, p2 being properties, C2-C4 being other classes. What needs to be added is the assertion that in both cases the same individual from C4 shall be referenced via p2. How can this be done?

 

Christian

 

 

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

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03

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

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03
Reply | Threaded
Open this post in threaded view
|

Re: Referencing the same individual in superclass expressions

Christian Boelling
Hi Matthew,

Matthew Horridge-2 wrote
If you make p1 and p2 functional this would work.  This might be too restrictive though.  Are C2, C3 arbitrary classes?
Indeed, this would be too restrictive. The question arose when trying to model (using BFO as upper level ontology) processes in which the continuant participants bear multiple roles that are realized in that process.

Example: In an elementary chemical reaction of type R, e.g. a complex formation, individual reactants of types A,B are transformed to a product of type C. I would like to be able to say that in each instance r of R the individual participants a and b of type A and B, respectively, each bear an input role and a component role. The component roles are - on top of the input roles - used to indicate the nature of the process, i.e. a non-covalent binding between the partners which maintain their identity in the complex. So the axioms for defining R would include:

realizes some (input_role and (has_bearer some A))
+
realizes some (component_role and (has_bearer some A))
+
a statement that in both these axioms the same instance from A must be referenced.

But that third statement seems to be difficult (or impossible) to express in OWL. This may seem like a very specific problem but to me it appears that in its general form, as posted originally, this issue could arise in a number of contexts.

Best

Christian