Re: Why this ontology is not marked as inconsistent?

Dear Mauro,
there is in fact no inconsistency (not in the open-world assumption that is
implied in OWL).
"W1" is a "Worker" by definition, but it is also a "Customer" for being
part of the range of the "hasCustomer" property.

To have some inconsistency you should say either that (e.g.):
1) "Customer" is disjoint with "Worker"; or
2) the class "Customer" contains exactly only the instance "C1"; or
3) the individual "W1" is not a "Customer".

In a "closed-world setting" you may expect 3) to be true because you do not
explicitly state "W1 is a Customer".
By the open-world assumption, instead, anything that you do not explicitly
negate can still be true.

Best,
Miguel

Il giorno sab 22 ago 2015 alle ore 12:51 Mauro Dragoni <dragoni@fbk.eu> ha
scritto:

> Dear all,
> I'm trying to model some inconsistency examples for working purposes but
> I'm not able to understand why this ontology is not marked as inconsistent
> by the Hermit reasoner through Protege.
>
> Briefly, I defined three concepts "Shop", "Customer", and "Worker".
> A property "hasCustomer" with domain "Shop" and range "Customer".
> Three individuals "S1", "C1", and "W1" of types "Shop", "Customer", and
> "Worker" respectively.
> On the "Shop" concept I defined a universal restriction on the
> "hasCustomer" property.
> Then, on the individual "S1", I instantiated the relationship
> "hasCustomer" with the individual "W1" (type Worker) instead of "C1" (type
> Customer).
>
> So, I expected an inconsistency message by the reasoner about the property
> range.
> Instead, it infers that "W1" is also of type Customer.
>
> I'm trying to figuring it out... but any help is appreciated.
>
> Kind regards,
> Mauro.
>
>
> --
> Dr. Mauro Dragoni
> Post-Doc Researcher at Fondazione Bruno Kessler (FBK-IRST)
> Via Sommarive 18, 38123, Povo, Trento, Italy
> Tel. 0461-314053
>
>