There are lots of papers published on successful applications of
neural net and neuro-fuzzy solutions sometimes extended with genetic
algorithms.  I am looking for discussions of problems associated with
these systems that make them ineffective or impractical to use.

---

Request for Question Clarification by tutuzdad-ga on 23 Jul 2003 08:23 PDT

Is this the type of discussion you are interested in?

EXAMPLE:

"Fuzzy Application Library"
http://www.fuzzytech.com/e/e_a_pfd.html
"Over the past few years, systems designers have been faced with a
rather controversial discussion about a new design technology named
"Fuzzy Logic". Some followers of fuzzy logic have pitched it as the
"magic bullet" for every type of engineering problem and have made
wild and unsupported claims in their publications. This has raised
skepticism amongst control engineering specialists and caused some to
completely condemn fuzzy logic as a marketing fad of no technical
value. "

Please let me know if I am on the right track.

Regards;
tutuzdad-ga

---

Clarification of Question by tmke-ga on 23 Jul 2003 10:37 PDT

I am looking for specific types of problems that neural nets and
neuro-fuzzy systems have had in application.  I 'assume' that there
have been problems 'training' neural nets.  I 'assume' that there may
be performance problems.  I 'assume' that these solutions 'may not'
respond to surprise information (information they have not been
trained to expect).  I 'assume' that the solutions of some of these
systems cannot easily be described.  I 'assume' that it is difficult
to develop flexible solutions using these technologies.  I am looking
for quotable problems that have been encountered while attempting to
use these technologies.

Hi Not much aware on the subject but have you tried stock market prediction :-)

The applications with which I am familiar include:

 1) Race books (horse racing analysis)
 2) Automated stock prediction
 3) Optical character recognition (at some point, the US Post Office
was/is using neural nets to read hand-written zip codes).

Cheers.