Re: intersectionOf and subClassOf

From: "Geoff Chappell" <geoff@sover.net>
Subject: RE: intersectionOf and subClassOf
Date: Sun, 30 Mar 2003 13:36:23 -0500

> > -----Original Message-----
> > From: www-rdf-logic-request@w3.org
> [mailto:www-rdf-logic-request@w3.org]
> > On Behalf Of Enrico Franconi
> > Sent: Sunday, March 30, 2003 10:51 AM
> > To: Geoff Chappell; www-rdf-logic@w3.org
> > Subject: Re: intersectionOf and subClassOf

[...]

> By "etc?", I meant that there seem to be an infinite number of
> equivalent descriptions of the same class. For example, aren't these all
> equivalent?
> 
> - A is a subClassOf B and C
> - A is a subClassOf a class that is the intersectionOf B and C
> - A is a subClassOf a class that is the intersectionOf B and a class
> that is the intersectionOf C
> - A is a subClassOf a class that is a subClassOf B and C
> - ...

These all appear to say the same thing.  (It is a bit difficult to state
this categorically because you are not being very precise here.)

> Would a reasoner be expected to decide that any one entails any of the
> others? Or just that they ultimately all result in the same subClassOf
> relationships on A? The former seems a difficult task for a rules-based
> reasoner (because you end up inferring an infinite chain of subClassOf
> relationships).

The relationship between OWL and RDF requires that there be comprehension
principles that require the existence of certain classes.  

For example, if there is an OWL class then there is a list containing only
this class and another class that is the intersection of the members of
this list.  If this intersection class is different from the first class,
it can start the same construction all over again.  There is nothing,
however, to prevent the intersection class from being the first class, but
there is also nothing to require it being the first class.

It is much better, however, to simply ignore all this machinery except as
is necessary for a particular entailment, where the machinery should do
just what is required to make the appropriate entailments happen.  

A forward-chaining reasoner is going to be tricky to build for OWL,
however, because of these comprehension principles.

peter

Received on Monday, 31 March 2003 08:46:32 UTC