ontolog-forum
[Top] [All Lists]

Re: [ontolog-forum] Axioms in Protege

To: "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
From: Patrick Cassidy <pcassidy@xxxxxxxxxxxxxxxx>
Date: Thu, 14 Aug 2003 13:17:47 -0400
Message-id: <3F3BC43B.8010308@xxxxxxxxxxxxxxxx>
Peter,
    The VNC viewer is amazing!  It works beautifully,
and could serve as a collaborative environment for developing
the ontology, if that is what this group would want.    (01)

    Adam is certainly correct (and I have said the same thing) that
the Ontology browser gives more complete information about
the SUMO ontology than the current port into Protege.  I have
no reason to duplicate that kind of effort, so I have not
attempted to make comprehensive references from each class
and slot to all of the axioms that mention them.  I view
the strength of Protege as providing a screen that summarizes
the main conceptual components of each class, giving one view
that can help understand the meaning of each class and its
primary relations to other concepts.
    At this early point where we are trying to organize the UBL
concepts into a formal ontology, I think that Protege can be
helpful in clarifying the basic components of each UBL
concept.  I find the SUMO browser also useful to search for
related concepts in SUMO.  To use that effectively, we
would need some method to collaboratively modify the
file (I presume it's a KIF file?) that the browser uses
as the knowledge base.
    If, after some experimenting, this group decides that using
Protege as the development environment is a good idea,
it will be necessary to create an accurate utility to
convert the Protege file into a KIF representation.
I haven't tried to do that and won't try in the near
future unless this group decides to use Protege as the
main development tool, or someone provides a business
application that uses a KIF file.    (02)

    Adam apparently doesn't consider the axioms in Protege to be a
"representation":
   [Adam pease] Protege can't represent SUMO's axioms.  The field into 
which you've entered them is essentially a comment field.    (03)

    I don't particularly care what one calls the axioms in Protege,
they can be accurately extracted and added to a KIF file.  There
are other potential problems in the Protege representation, such
as the lack of direct support for instances of multiple classes.
There is a work-around, but I haven't looked at it carefully yet
to determine if it would accurately reproduce a KIF file used
as the canonical representation.
    At this point one thing that I have not considered in
using Protege is how to organize the concepts within
"modules" or "microtheories".  For the Invoice ontology, we
probably can view that as a single microtheory, so this
issue is not likely to be a problem at this early stage.    (04)

     Pat    (05)



Peter P. Yim wrote:    (06)

> 
> 2. Once again, I've hoisted your work over to the VNC server (which we 
> can use for our meeting(s), and for people who don't have a Protégé 
> setup on their computer to view and play around with.
> 
=============================================
Patrick Cassidy    (07)

MICRA, Inc.                      || (908) 561-3416
735 Belvidere Ave.               || (908) 668-5252 (if no answer)
Plainfield, NJ 07062-2054        || (908) 668-5904 (fax)    (08)

internet:   cassidy@xxxxxxxxx
=============================================    (09)

_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Unsubscribe/Config: 
http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx    (010)

<Prev in Thread] Current Thread [Next in Thread>