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] re asoner doesn't react to intersection

bassiee back_to_basic_69 at
Tue May 6 17:21:01 PDT 2008

Sorry to bother again,
but I have another similar question:

rule in classA:
(= 1 hasA owl:Thing) ⊔ (= 1 hasB owl:Thing) 
 (extra information: hasA, hasB are both connect with Domain ClassA and
Range ClassB)

so the rule says that one of both properties should have only one resource.
So in ClassA I make an instance and:
the property of 'hasA' doesn't have any resources 
and 'hasB' has for example 5 resources
the reasoner doesn't see this as a problem due the open world assumption
(because there can still be one resource added to the property 'hasA' so the
reasener doesn't see this ass a problem)
This part I understand.

2/I add 2 resources to the property of 'hasA' and 2 resources to the
property of 'hasB' 
(one of both properties should have only one resource)
The reasoner doesn't react to it. 
This worries me...
Is this normal?
At this moment I'm testing with pellet and racer pro using the dig
If it is normal that the reasoner doesn't react to it, is there another way
to test it?

Thanks in advance, and thanks for all your help so far.

Thomas Russ wrote:
> On May 6, 2008, at 10:18 AM, bassiee wrote:
>> So due the open world reasoning the reasoner doesn't see this ass a  
>> problem,
> Correct.  It is not a problem.
>> but in my ontologie there should be at least one.
>> Must I change the rule, or is it correct?
> The rule is correct.  You have properly stated the requirement that  
> there be such a filler.  OWL knows that there is such a filler.  It  
> just doesn't happen to know exactly which individual is the filler.
> Since you have a correct model, there isn't anything you can really  
> change.

View this message in context:
Sent from the Protege - OWL mailing list archive at

More information about the protege-owl mailing list