[xtm-wg] analysis of the meaning of ontology

["Paul Prueitt" <beadmaster@ontologyStream.com>]

I am reading Waterson and Preece's "Verifying Ontology Commitment in
Knowledge-Bases Systems"  which was recommended either here or in the XTM-WG
forum.  I printed the copy out from the Internet, but as it was pdf, the URL
address was not printed.

The paper is excellent, and yet there is an assumption that any "knowledge
base" must really be a expert system type structure.  In spite of this, the
issues raised regarding certification of ontology are relevant to my own
work on Frank Sowa Peircean type cognitive graphs.

In the Waterson and Preece paper the perception is that ontologies are
supplemental to the rule set in the modern expert system, and that if one
considers the merge of two knowledge bases, then one needs to also consider
how to combine the ontologies.  They nicely define an ontology to be "An
ontology defines the terminology of a domain of knowledge: the concepts that
constitute the domain, and the relationships between those concepts."

The rules (expert system) by which knowledge is made relevant is a somewhat
separate system.

First, I wish to recognize that there are really two difference schools of
thought.  One school feels that knowledge can be represented as rule
structures, ontologies and languages for communicating and expressing the
rules.  This clearly has value in specific context.

A second school of thought feels that knowledge is something experienced and
that representation as rules is not always possible.  The ontology remains
viable and useful without the expert system, since the humans in the system
can relate to the ontology and use natural intelligence to do constructive
and productive work.  Moreover the ontology itself can be used as an
catalyst in routing and retrieval.

This does not mean that rules in work flow does not make some sense, clearly
work flow rules do make sense.

If one gives up the dependency on systems of rules, and just focus effort on
the ontology, then one can start to realize semiotic-type systems (system of
language like symbols) that produce situational ontologies where the
concepts (location) and the linkage between the concepts are formative.
This leaves the expert system type rules to the research lab, well almost.
It leaves the rules to high level analytic processes that give global
guidance.

Does someone have the URL link to the Waterson and Preece paper?

What I would like to do is to develop a discussion about the second school
of thought and how this second school of thought might address the
commercial need to have a product.  The discussion of the Acappella
Innovation might also be appropriate if everyone agrees that this is of
interest.





------------------------ Yahoo! Groups Sponsor ---------------------~-~>
eGroups is now Yahoo! Groups
Click here for more details
http://click.egroups.com/1/11231/0/_/337252/_/980730345/
---------------------------------------------------------------------_->

To Post a message, send it to:   xtm-wg@eGroups.com

To Unsubscribe, send a blank message to: xtm-wg-unsubscribe@eGroups.com