Re: TEST, SEM: test cases for dark triples

> > [...]
> >
> > > > OWL-entails
> > > >
> > > >   :John a ( owl:Restriction _:x owl:onProperty :child owl:hasClass _:x )
.
> >
> > [...]
>
> > the () are indeed regular owl lists
>
> But then :John becomes an instance of the list given above.
> This is not a valid consequence, or, at least, does not appear to me to be
> a valid consequence in any version of OWL/FOWL/DAML+OIL that I am familiar
> with.
>
> Are you indeed deducing that :John is a member of an object that is an
> instance of daml:List?  If so, please let me know what sort of logical
> system you have implemented in your rules.

At that time we *had* entailment rules like

{ ?s ?p ?o . ?s a ( owl:Restriction ?R owl:onProperty ?p owl:toClass ?C ) }
  log:implies { ?o a ?C } .

{ ?s ?p ?o . ?o a ?C }
  log:implies { ?s a ( owl:Restriction ?R owl:onProperty ?p owl:hasClass ?C ) }
.

That worked as such but was meaningless/misleading and is now thrown away.
We now have (as a matter of test) another possible pair of er

{ ?s ?p ?o . ?s a [ owl:restrictionOf ( ?p ?C ) ] } log:implies { ?o a ?C } .

{ ?s ?p ?o . ?o a ?C } log:implies { ?s a [ owl:restrictionOf ( ?p ?C ) ] } .

and for instance on the question wether

  :John :child :Bill .
  :Bill a :Person .

owl-entails

  :John a _:1 .
  :John a [ owl:restrictionOf ( :child _:1 ) ] .

we get an positive answer, for instance Euler found

  :John a [ owl:restrictionOf ( :child [ owl:oneOf ( :John :Bill ) ] ) ] .

--
Jos

--