Reasoning with Fact++ can not finish.

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

Reasoning with Fact++ can not finish.

Xiaojuan
Hi, there,
I found that if I use restrictions 'exactly 1' or 'value' in the ontology,
the reasoning with Fact++ would be very slow. If there are many these kind
of restrictions, the reasoning could not finish. Is there anyone who knows
how to improve this? Here is the information after I ran the reasoning:

'''
Pre-computing inferences:
    - class hierarchy
    - object property hierarchy
    - data property hierarchy
    - class assertions
    - object property assertions

[GitRepo] Git repository detected: /home/xhu/github/aadbonto/.git
[GitRepo] On branch: master
'''
Then the protege just like dead, I have to restart. And sometimes it died
here:

'''
Pre-computing inferences:
    - class hierarchy
    - object property hierarchy
    - data property hierarchy
    - class assertions
    - object property assertions

'''

Thank you in advance!



--
Sent from: http://protege-project.136.n4.nabble.com/Protege-User-f4659818.html
_______________________________________________
protege-user mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-user
Reply | Threaded
Open this post in threaded view
|

Re: Reasoning with Fact++ can not finish.

Lorenz Buehmann
Any single OWL axiom of any kind can make the reasoning from being
tractable to almost unusable. That's the point where people tend to try
other reasoners, each of them has different advantages and optimizations.

On 21.04.20 12:22, Xiaojuan wrote:

> Hi, there,
> I found that if I use restrictions 'exactly 1' or 'value' in the ontology,
> the reasoning with Fact++ would be very slow. If there are many these kind
> of restrictions, the reasoning could not finish. Is there anyone who knows
> how to improve this? Here is the information after I ran the reasoning:
>
> '''
> Pre-computing inferences:
>     - class hierarchy
>     - object property hierarchy
>     - data property hierarchy
>     - class assertions
>     - object property assertions
>
> [GitRepo] Git repository detected: /home/xhu/github/aadbonto/.git
> [GitRepo] On branch: master
> '''
> Then the protege just like dead, I have to restart. And sometimes it died
> here:
>
> '''
> Pre-computing inferences:
>     - class hierarchy
>     - object property hierarchy
>     - data property hierarchy
>     - class assertions
>     - object property assertions
>
> '''
>
> Thank you in advance!
>
>
>
> --
> Sent from: http://protege-project.136.n4.nabble.com/Protege-User-f4659818.html
> _______________________________________________
> protege-user mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-user
_______________________________________________
protege-user mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-user
Reply | Threaded
Open this post in threaded view
|

Re: Reasoning with Fact++ can not finish.

Xiaojuan
Thank you for your reply! It turns out that the reasoning can be done on my
colleague's laptop with Fact++. Could it be a problem of settings of
Protege?



--
Sent from: http://protege-project.136.n4.nabble.com/Protege-User-f4659818.html
_______________________________________________
protege-user mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-user
Reply | Threaded
Open this post in threaded view
|

Re: Reasoning with Fact++ can not finish.

samsontu


On Apr 21, 2020, at 4:34 AM, Xiaojuan <[hidden email]> wrote:

Thank you for your reply! It turns out that the reasoning can be done on my
colleague's laptop with Fact++. Could it be a problem of settings of
Protege?


Check the Java heap size allocation for the Protege process (the -Xmx parameter). Limited memory is a possible bottleneck for any Java process.

With best regards,
Samson


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