[Pellet-users] is there a way to see how Pellet decides for an entailment?

[vasilis spiliopoulos]

On Dec 8, 2007 1:18 PM, vasilis spiliopoulos <2billy at gmail.com> wrote:

> Thanks for the help!! :-)
>
>
> On Dec 8, 2007 5:06 AM, Evren Sirin <evren at clarkparsia.com> wrote:
>
> > On 12/5/07 11:22 AM, vasilis spiliopoulos wrote:
> > >
> > > On Dec 5, 2007 5:50 AM, Evren Sirin <evren at clarkparsia.com
> > > <mailto:evren at clarkparsia.com>> wrote:
> > >
> > >     On 12/4/07 5:13 AM, vasilis spiliopoulos wrote:
> > >     > Hi List!
> > >     >
> > >     > Is there a way to see how Pellet decides for an entailment?
> > >
> > >     Well, it depends on what you mean by "seeing" :) Pellet is a
> > tableau
> > >     reasoner and uses a refutation technique, e.g. adds the negation
> > >     of what
> > >     you are trying to prove and checks for inconsistency. By setting
> > the
> > >     logging level to debug you can see verbose messages showing when a
> > >     tableau rule fire, when a clash occurs and when backtracking
> > happens.
> > >     This log requires some understanding of how tableau algorithm work
> > (at
> > >     least reading chapter 2 in Description Logic Handbook).
> > >
> > >     Pellet also provides explanations for an inference in the form of
> > >     a set
> > >     of axioms that cause the inference. For example, see the
> > >     ExplanationExample program that comes in the distribution package.
> > One
> > >     need to know the semantics of OWL axioms and constructs to make
> > the
> > >     connection between the explanation set and the inference.
> > >
> > >
> > > I am using jena + pellet. Is it possible to get such explanations
> > > through jena?
> > > I just do:
> > > OntModel model =
> > > ModelFactory.createOntologyModel(PelletReasonerFactory.THE_SPEC );
> > >
> > >
> > >
> > >     >
> > >     > For the specific concepts of a loaded ontology or theoretically
> > in a
> > >     > manual (in the form of examples).
> > >
> > >     I'm not sure what you mean here. There are various sources
> > describing
> > >     how tableau algorithms work (again I refer to chapter 2 of DLHB)
> > which
> > >     might help.
> > >
> > >
> > > For example: We have to ontologies and make a union of these two
> > > ontologies. Then we state some equivalences between e.g. properties or
> > > classes and then we ask pellet for relations between classes coming
> > > from the two different initial ontologies. Such examples. Why pellet
> > > states that concept a is less than concept b. Is there some "simple
> > > and intuitive" rules? No tableau algorithms.
> >
> > Try the explanation example I mentioned. For any kind of entailment
> > (subclass of, instance of, etc.) it will give you the set of asserted
> > axioms that caused the inference. It doesn't tell you exactly how those
> > axioms cause the inference but just seeing those couple of axioms that
> > are filtered out of many axioms in the ontology helps a lot. You can try
> >
> > the explanation feature through OWSight [1] too
> >
> > Cheers,
> > Evren
> >
> > [1] http://pellet.owldl.com/ontology-browser/
> >
> > > I do not know if I am clear enough
> > >
> > >
> > >
> > >     Cheers,
> > >     Evren
> > >
> > >
> > > Thanks for the reply!!
> > >
> > >
> > >
> > >     >
> > >     > Thanks!!
> > >     >
> > >     > --
> > >     > vassilis
> > >     >
> > >
> > ------------------------------------------------------------------------
> > >     >
> > >     > _______________________________________________
> > >     > Pellet-users mailing list
> > >     > Pellet-users at lists.owldl.com <mailto:Pellet-users at lists.owldl.com
> > >
> > >     > http://lists.owldl.com/mailman/listinfo/pellet-users
> > >     > _______________________________________________
> > >     >
> > >     > Sponsored by Clark & Parsia, LLC http://clarkparsia.com/
> > >
> > >
> > >
> > >
> > > --
> > > vassilis
> >
> >
>
>
> --
> vassilis




-- 
vassilis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.owldl.com/pipermail/pellet-users/attachments/20071209/3ba09e43/attachment.htm