constructs that make reasoning slow

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

constructs that make reasoning slow

Daniel Schober
Dear DL reasoning experts,

I wonder if there are investigations in what constructs in OWL 2 RL are particulary reasoning intensive / time costly for DL reasoners like HermIT, Pellet and RACER.

The current expressivity of our ontology (DCO) is SRIF(D). Hermit already takes ~2 minutes to classify DCO including BioTop on an average PC.


Could anyone point me to a list of things that can be simplified in an owl DL ontology to make reasoning faster ? Ideally it would be some guideline that can be understood by non-DL experts (ontology engineers in the biomedical domain) as well ;-) .

The table illustrates the statistics of DCO and its imported BioTop level ontology:

Ontology elements and axioms

Count (all)

DCO

BioTop

Classes

1281

965

375

Object Properties (relations)

78

3

74

Datatype Properties

11

10

0

Subclass Axioms

1494

1050

444

Equivalent Class Axioms

197

98

99

Disjoint Axioms

76

1

75



Cheers, Daniel Schober.
-- 

-------------------------------------------------------------
  Dr. Daniel Schober
  Universit├Ątsklinikum    
  Institut f├╝r Medizinische Biometrie und Medizinische Informatik
  Stefan-Meier-Strasse  26	Raum 6
  D-79104 Freiburg		Germany
  Tel: +49 (0)761 2036807	FAX: +49 (0)761 2036711
  
  
          
          
          
          
             
          
          
          

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03
Reply | Threaded
Open this post in threaded view
|

Re: constructs that make reasoning slow

gfrishkoff

Dr. Daniel Schober <schober <at> imbi.uni-freiburg.de> writes:

>
>
> Dear DL reasoning experts,
> I wonder if there are investigations in what constructs in OWL 2 RL are
> particulary reasoning intensive / time costly for DL reasoners like
> HermIT, Pellet and RACER.

Dear Dr. Schober,

Did you ever receive an answer to your query below? I have the same question.
I've added some content to our ontology in the last couple of days and seen an
exponential increase in the time it takes HermiT to classify.

I'd also like to know how to optimize performance of HermiT (or Pellet) or where
to find a faster reasoner, since it's not an option (currently) to remove
anything from the ontology.

Thanks,
Gwen Frishkoff
Assistant Professor of Psychology & Neuroscience
Georgia State University
[hidden email]



_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03
Reply | Threaded
Open this post in threaded view
|

Re: constructs that make reasoning slow

Timothy Redmond

There are some experts on this list who might be able to give you an
answer to this question but I suspect you will have better luck
elsewhere.  Perhaps public-owl-dev mailing list (http://lists.w3.org)?

-Timothy


On 01/13/2011 10:08 PM, gfrishkoff wrote:

> Dr. Daniel Schober<schober<at>  imbi.uni-freiburg.de>  writes:
>
>>
>> Dear DL reasoning experts,
>> I wonder if there are investigations in what constructs in OWL 2 RL are
>> particulary reasoning intensive / time costly for DL reasoners like
>> HermIT, Pellet and RACER.
> Dear Dr. Schober,
>
> Did you ever receive an answer to your query below? I have the same question.
> I've added some content to our ontology in the last couple of days and seen an
> exponential increase in the time it takes HermiT to classify.
>
> I'd also like to know how to optimize performance of HermiT (or Pellet) or where
> to find a faster reasoner, since it's not an option (currently) to remove
> anything from the ontology.
>
> Thanks,
> Gwen Frishkoff
> Assistant Professor of Psychology&  Neuroscience
> Georgia State University
> [hidden email]
>
>
>
> _______________________________________________
> protege-owl mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-owl
>
> Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03