Search Mailing List Archives
[protege-owl] Negated Properties in SWRLTab
std01092 at di.uoa.gr
Wed Aug 15 02:38:19 PDT 2007
I was just referring to NAF to demonstrate the difficulty in defining
negation in SWRL at least for the lhs of the rules.
There's only the ComplementOf construction which could work as unary
negation over classes.
Moreover, if you're talking about the rhs, a simple workaround is to
build a class in OWL that satisfies the desired properties (or their
negation) and have the SWRL rule assert the variable object as an
instance of that class.
Matt Williams wrote:
> Dear Petros,
> Thanks for that - just a point that I was referring to negation, not
> NAF, and they are not the same - after all, OWL has negation, but does
> not have NAF.
> Having said that, I can't see any references in the SWRL docs that
> talk about negation, but then that would mean we can't write:
> A(x) -> not B(x)
> which would seem to be a fairly trivial assertion to make (and since
> it does not extend OWL, should also not cause problems with reasoning).
> Papapanagiotou Petros wrote:
>> Dear Matt,
>> I'm afraid negation is not supported by SWRL.
>> Adding support for negation isn't a trivial matter.
>> Negation-as-failure is contradictory to the SW open-world assumption.
>> Thus, SWRL is, unfortunately, limited in this regard.
>> You'll have to think of a way around it, if any.
>> Matt Williams wrote:
>>> Dear All,
>>> Is there any way of using negated properties in SWRL?
>>> I would like to use a rule such as:
>>> ?a myProp ?b -> ?a not myOtherProp ?b
>>> but the editor won't allow me.
>>> Any ideas?
>> protege-owl mailing list
>> protege-owl at lists.stanford.edu
>> Instructions for unsubscribing:
More information about the protege-owl