Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[protege-dev] Ontology Performance

Lorenz B. buehmann at informatik.uni-leipzig.de
Thu Feb 23 00:19:02 PST 2017



> Thank you Sir Csongor, 
> In fact, I do some research and I found that some works check the effect of
> the swrl number in the performance by calculating the execution time. 
> I hope that you don't forget my past posts, my ontology is Big. for that I
> have idea to check the effect of the number of individual in the performance
> of ontology but I don't have how to do this. 
> In other hand I found some research I found also some metrics used to check
> the complexity of ontology. These metrics are: 
> Paths: (the sum of path in ontology) 
> X: the max path length 
> Y: the average path length 
> D: average path per concept 
> Z: average relationship per concept
> How may I calculate these metrics Please. 
> I hope that you have idea about this two questions 
> Best regards 
Define "path" in an ontology first. Then use OWL API, JENA API or
whatever and do in Java.
By the way, there is no proper way of measuring the reasoning
performance of an ontology as it also depends on the reasoner.
This is research - a good thesis about it is definitely [1]

 [1] www.cs.man.ac.uk/~sazonauv/SazonauThesis.pdf
>
>
>
> --
> View this message in context: http://protege-project.136.n4.nabble.com/Ontology-Performance-tp4667640p4667645.html
> Sent from the Protege Developer mailing list archive at Nabble.com.
> _______________________________________________
> protege-dev mailing list
> protege-dev at lists.stanford.edu
> https://mailman.stanford.edu/mailman/listinfo/protege-dev
>
-- 
Lorenz Bühmann
AKSW group, University of Leipzig
Group: http://aksw.org - semantic web research center

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/protege-dev/attachments/20170223/1aefe76b/attachment.html>


More information about the protege-dev mailing list