[Pellet-users] Slow inference/Restrict inference

[Alejandro Rodríguez González]

Hi Serge,

When i'm finish my february exams and i have a little more of free time 
i will try to make some test removing axioms.

The reason of this axioms was because it makes the "optimium" reasoning, 
because give the exact inference result that i was searching. The 
problem is that the reasoning time is too high so.. it's not applicable.

I was thinking that maybe another option could be, as you said, remove 
one kind of axioms, and refine the results programatically.. Anyway, i 
will try to search other axioms that could improved the speed inference.

I will keep you informed.

Thanks!

PD: Kendall, i want to apologize too if the way i'm expressed myself was 
wrong. I understand that you make a hard work in pellet developer and 
that you ask questions when you can. Sorry :-)

Serge Libotte escribió:
> Hola Alejandro,
>
> I'm not suggesting that your modeling approach is wrong.
> Just give a try to some of the recommendations I've made to test their 
> impact on the behaviour of the reasoner. I found that just removing 
> just one kind of axiom can make things technically better (or at least 
> technically possible) If it's the case then youre are facing a 
> modeling limitation because of the technical implementation. This 
> could be frustrating but is not unusual and you then have to find a 
> work around.
>
> Be kind to keep me informed about any conclusion you have as I'll do 
> on my side the same kind of work.
>
> Regards,
>
> Serge.
>
>
> 2008/2/1, Alejandro Rodríguez González <jalo.javier at gmail.com 
> <mailto:jalo.javier at gmail.com>>:
>
>     Hi Serge,
>
>     In first place, thanks for your answer :-)
>
>     I was checking my ontology and my code. I have UNA option set tu true,
>     and talking about axioms..
>
>     This is an example of the axioms that i have in my classes:
>
>     (hasSymptom some Nausea_and_vomiting_Class) or (hasSymptom some
>     Fever_unspecified_Class) or (hasSymptom some Headache_Class) or (....)
>
>     And
>
>     hasSymptom only (Nausea_and_vomiting_Class or
>     Fever_unspecified_Class or
>     Headache_Class or ...)
>
>     This is to specify that the disease can have the mentioned
>     symptoms, and
>     only the mentioned symptoms.. Cardinality axioms are only set in the
>     "test instances".
>
>     This is made because RDF/OWL make an open world assumption, so, if
>     i say
>     that the "test instance" has symptoms A,B and C that doesn't mean it
>     doesn't also have other symptoms not yet mentioned in the description.
>     So, adding a cardinality restriction of 3 on the individual and
>     forcing
>     all the symptom instances to be distinct i'm clossing the world and
>     making it possible to test compliance with the allValuesFrom
>     restriction.
>
>     So.. i'm not sure what to do, because if i change this axioms probably
>     the inferences will be wrong.. :-/
>
>     Serge Libotte escribió:
>     > Hi Alejandro,
>     >
>     > Strangly this is the kind of question the list tend to forget it
>     seems ;-)
>     >
>     > Try those kind of things:
>     > - Set UNA option of Pellet to 'true'
>     > - Remove ALL 'one of' axioms
>     > - Remove ALL cardinality Axioms
>     >
>     > Those recommendations were inspired by reading carefully the
>     attached
>     > document.
>     >
>     > If it helps, wonder if you can live without those axioms.
>     > On our side we are testing other reasoners.
>     >
>     > Good luck.
>     >
>     > Serge.
>     >
>     >
>     >
>     > 2008/1/31, Alejandro Rodríguez González <jalo.javier at gmail.com
>     <mailto:jalo.javier at gmail.com>
>     > <mailto:jalo.javier at gmail.com <mailto:jalo.javier at gmail.com>>>:
>     >
>     >     Hello,
>     >
>     >     I ask a few days ago about how to restrict the inference
>     domain in
>     >     pellet but no one answered me so.. i will try to make again the
>     >     question
>     >     (maybe no one understand me, i don't know).
>     >
>     >     I have an ontology (
>     http://www.jalojavier.es/humandisease.owl ) and
>     >     Jena+pellet code ( http://rafb.net/p/k1yHie11.html ) to make the
>     >     inferences.
>     >
>     >     The problem is that the inference that i try to make with
>     the code
>     >     mentioned take a lot of time (near to 180 seconds)..
>     >
>     >     I make some test splitting the ontology into small parts,
>     and i think
>     >     that the problem is the number of instances of the ontology
>     (now i
>     >     have
>     >     near to 600 individuals and 1500 classes)..
>     >
>     >     I think that may be it's possible to optimize the inference
>     speed
>     >     making
>     >     a restriction over the inference domain. I have "Diaseses",
>     >     "Symptoms",
>     >     and "Lab Test" superclasses that are involucrated in the
>     inferences,
>     >     but, the results of the inferences are only subclasses of
>     "Diseases"
>     >     (that only has near to 30 instances).
>     >
>     >     It's possible say to pellet that only must search the
>     results in this
>     >     superclass and ignore the rest?
>     >
>     >     Or any other solution that optimize the inference speed..
>     >
>     >     Thanks.
>     >     _______________________________________________
>     >     Pellet-users mailing list
>     >     Pellet-users at lists.owldl.com
>     <mailto:Pellet-users at lists.owldl.com>
>     <mailto: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/
>     >
>     >
>
>