Dear Ed, (01)
> Matthew West wrote:
>
> > [MW] The first thing I would do is ask for a review of the
> models/ontologies
> > developed. There are a number of quality checks that can be done that
> are
> > independent of the subject area, the results of which can improve the
> > utility of the models/ontologies.
>
> Since Howard avers that the community of domain experts has reviewed
> and
> approved these models as accurate knowledge, I assume that Matthew is
> suggesting some kind of "model quality" analysis. If there is
> published
> agreement on best practices in modeling (by a large body of experts in
> that field), then a team can be trained to employ those in doing the
> model analyis. But I'm not aware of any such published agreement, and
> the last thing the systems engineering models need is model analysis by
> a self-appointed team with a consistent but not clearly appropriate
> viewpoint. (02)
[MW] There are actually at least a couple of approaches you could try: (03)
- Nicola Guarino and Chris Welty have come up with a series of quality
analysis you can perform (Ontoclean)
- I would of course suggest a "High Quality Data Models" analysis.
- You also have the Boro methodology.
>
> Howard, Matthew and others will recall the damage done to well-
> developed
> domain models by a team of expert SQL database designers considered to
> be "information modeling" experts in a not-so-distantly related ISO
> activity. I can envisage coercion of the models to OWL/DL by
> well-meaning model analysts who have a similar shortage of breadth in
> their vision. (04)
[MW] Well if you accept the advice of self appointed experts without
critical analysis, then you deserve what you get. (05)
[MW] Speaking of coercion to OWL is relevant to ISO 15926. There have been
several attempts at this now, with different priorities given to being
truthful to the original expressivity of the ISO 15926 data model, or
maximising the ability to use OWL reasoning features at the expense of
expressiveness (or both). However, I'm glad to say that the debate is
actually rather well informed about what exactly is at stake.
>
> So the question is: Where do the model analysis guidelines come from? (06)
[MW] See above for some suggestions. (07)
> And how will the analysts work with the domain experts to understand
> the
> intent when they don't like the structure of the ontology? (08)
[MW] Well that is the key to success, I think that (good) ontologists would
be able to through up things that look suspicious by looking at the
structure of the ontology, but the "What did they really mean?" question has
to go back to the domain experts, perhaps with a few options to prompt them. (09)
Regards (010)
Matthew West
http://www.matthew-west.org.uk/ (011)
>
> -Ed
>
> --
> Edward J. Barkmeyer Email: edbark@xxxxxxxx
> National Institute of Standards & Technology
> Manufacturing Systems Integration Division
> 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528
> Gaithersburg, MD 20899-8263 FAX: +1 301-975-4694
>
> "The opinions expressed above do not reflect consensus of NIST,
> and have not been reviewed by any Government authority."
>
> _________________________________________________________________
> Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
> Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
> Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
> Shared Files: http://ontolog.cim3.net/file/
> Community Wiki: http://ontolog.cim3.net/wiki/
> To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
> To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
> (012)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (013)
|