Re: OWL - restrict subclassing?

Heiko Gottschling wrote:

> On Monday 08 December 2003 14:06, jon@hackcraft.net wrote:
> 
> 
>>I think that's okay, but why not have:
>>
>><owl:Class rdf:ID="Fruit">
>>    <owl:unionOf rdf:parseType="Collection">
>>        <owl:Class rdf:resource="#Apple" />
>>        <owl:Class rdf:resource="#Banana" />
>>    </owl:unionOf>
>></owl:Class>
> 
> 
> That brings me to another question - given the above definition of fruit, 
> would the following be valid (or even implied?):
> 
> <owl:Class rdf:ID="Apple">
>   <rdf:subClassOf rdf:resource="#Fruit" />
> </owl:Class>
> 
> IAW, what is the relation between 'unionOf' and 'subClassOf'? 

If class A is defined as the union of classes B and C, then one can 
derive that B and C are subclasses of A (the class extensions of B and C 
must be subsets of the class extension of A).
Similarly, if class A is defined as the intersection of classes B and C, 
then one can derive that A is a subclass of both B and C.

> Would it be ok 
> to say that a class represents the union of its subclasses?

No, unless explicitly defined otherwise, OWL subclass partitions may be 
incomplete and/or overlapping.

Guus


> 
> Heiko
> 

-- 
Free University Amsterdam, Computer Science
De Boelelaan 1081a, 1081 HV Amsterdam, The Netherlands
Tel: +31 20 444 7739/7718
E-mail: schreiber@cs.vu.nl
Home page: http://www.cs.vu.nl/~guus/

Received on Monday, 8 December 2003 09:20:40 UTC