The process of extracting knowledge from an expert and coding it into
an expert system (ES)is the domain of the "knowledge engineer" (KE,) a
hardworking individual who must be a programmer, but also endeavor to
learn the domain knowledge from the expert(s). In recent years, there
have been numerous attempts to simulate the work of the KE in
software, by providing a friendly authoring GUI for an ES, or some
kind of natural-language-like rule language, or both. Many commercial
products sport these features. Jess, being a research tool, does not
offer either of these concessions to "user-friendliness." No
pretensions are made that anyone other than a knowledgable programmer
should be able to create an ES with Jess.
AFAIK, it turns out that these user-friendliness tools are not
sufficient, in any case, to allow non-programmer domain experts to
build an ES by themselves. Most if not all real ESs must still be
partially or completely built by a KE working together with the
experts. The vendors of the commercial systems will certainly tell you
otherwise, and rather than listening to me, it would be best to get
the advice of an experienced user of such a system before buying
one.
You may be able to have a human KE work with the expert for a short
time, accomplishing two things: first, laying the groundwork for the
system (discovering the important abstractions, choosing data
representations, finding some of the top-level rules,) and in the
process, teaching the expert to do some programming on her own.
I think Peter Frederick wrote:
[Charset iso-8859-1 unsupported, filtering to ASCII...]
> Dear All,
> I am planning on using JESS for developing an expert system for use in an
> e-commerce software package. I need to capture the knowlege from a group of
> potential users who have no knowledge of JESS.
> If someone has been in a similar situation, please could you give some
> advice as to how best to get the experts to put their knowledge into a
> knowledge base/rules (format and structure) as I have had some difficulty
> trying to put together my own 'how-to' from the manual.
>
> Yours Sincerely,
>
> Peter Frederick
>
---------------------------------------------------------
Ernest Friedman-Hill
Distributed Systems Research Phone: (925) 294-2154
Sandia National Labs FAX: (925) 294-2234
Org. 8920, MS 9012 [EMAIL PROTECTED]
PO Box 969 http://herzberg.ca.sandia.gov
Livermore, CA 94550
---------------------------------------------------------------------
To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the
list (use your own address!) List problems? Notify [EMAIL PROTECTED]
---------------------------------------------------------------------
