RE: RDF and OWL rules

Hi Jos,

I'm struggling a bit with one of the test cases [1]. It makes intuitive
sense to me but I can't quite see how it works according to the
semantics doc [2]. I'm hoping you might be able to point out where I'm
going wrong in my reasoning.

In this test case, the premise is a cardinality restriction on a
property. The conclusion is that the restriction in the premise is
identical to an intersection of an equivalent min and max cardinality on
the same property.

I can see how the conclusion could say that the restriction C in the
premise is an equivalentClass to some class that is an intersectionOf
the two new restrictions (i.e. they have the same class extension), but
I'm having trouble seeing what licenses one to say that they (the two
classes) are one and the same.

Perhaps there's something wrong with my understanding of the semantics
of intersectionOf? My understanding is that they are if, not iff. I
thought that meant that we can't infer that C is an intersection of A
and B just because we know that the class extension of C is the same as
the intersection of the class extensions of A and B. (this confuses me
somewhat because I thought I'd received contradictory replies to a
similar query on rdf logic[3] - but perhaps I was too imprecise in what
I asked)

I hope I've managed to explain myself :-)

Regards,

Geoff

P.S. I think that rule4p1 is, strictly speaking, unnecessary since: 

owl:equivalentProperty rdfs:subPropertyOf rdfs:subPropertyOf; a
owl:SymmetricProperty;

(as long as the standard rdfs rules are included).


[1] http://www.w3.org/2002/03owlt/cardinality/Manifest005

[2]
http://www.w3.org/TR/2003/WD-owl-semantics-20030331/rdfs.html#owl_AllDif
ferent_rdf

[3] http://lists.w3.org/Archives/Public/www-rdf-logic/2003Mar/0185.html

Received on Saturday, 5 April 2003 22:02:19 UTC