Loading ...
Sorry, an error occurred while loading the content.

RE: [TaxoCoP] OWL2 Engines

Expand Messages
  • Jim Tivy
    Andrea Thanks for the list of triple stores. I will have a look. We tried Jena but found its transaction model difficult to integrate with our parent
    Message 1 of 3 , Oct 28, 2011
    • 0 Attachment

      Andrea

       

      Thanks for the list of triple stores.  I will have a look.

       

      We tried Jena but found its transaction model difficult to integrate with our “parent” transaction – eg: we need to do some operations in the CMS and in the triple store in the same transaction (would rather not use 2 phase commit).  As well, Jena does not cleanly model the first class objects of OWL2 – classes,individuals,properties.  It was not obvious how to update the ontology aside from loading an updated file.  File imports did not support the OWL2 Functional Language.  As well, inference is used to determine class descendancy whereas a cached graph is more straight forward.

       

      So agreed, a triple store does the work underneath, but we need a clean API on top of that with good plugin transaction support.

      We are thinking of just writing one – we see it as about 6 tables to start.

       

      Jim

       

      From: TaxoCoP@yahoogroups.com [mailto:TaxoCoP@yahoogroups.com] On Behalf Of Andrea Splendiani
      Sent: October-28-11 1:32 AM
      To: TaxoCoP@yahoogroups.com
      Subject: Re: [TaxoCoP] OWL2 Engines

       

       

      Hi,

       

      you don't need and OWL2 engine for that. A triplestore which supports minimal inference is ok. Some people even materialize inferred statements on the triplestore, and can avoid doing reasoning on demand.

      There is a range of freely available triplstores: Virtuoso, 4store, Jena, BigOWLIM (one variant is free) and others!

       

      best,

      Andrea

       

      Il giorno 28/ott/2011, alle ore 04.37, Jim Tivy ha scritto:



       

       

      We are looking for an OWL2 storage engine – likely relational based.  Something simple, either free or with an affordable OEM license.  We are not so much interested in doing inferences but rather want to store classes and properties and perform fairly simple queries on them such as;


      Select all the individuals in class A or class A and descendants.

      Select all the individuals with property Q of value “Red” or individual 502.

       

      ___

       

       

      Andrea Splendiani

      Senior Bioinformatics Scientist

      Centre for Mathematical and Computational Biology

      +44(0)1582 763133 ext 2004

       

       

       

    Your message has been successfully submitted and would be delivered to recipients shortly.