Newsgroups: comp.ai.fuzzy
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel-eecis!news.mathworks.com!uunet!in2.uu.net!EU.net!usenet2.news.uk.psi.net!uknet!usenet1.news.uk.psi.net!uknet!psinntp!psinntp!psinntp!mercury.hsi.com!news
From: mascarton@wpmail.code3.com (Mark A. Scarton)
Subject: Re: Extracting rules from neural network
X-Nntp-Posting-Host: pc027064.code3.com
Message-ID: <31fcdd0c.428350475@mercury.hsi.com>
Sender: news@hsi.com (USENET news)
Reply-To: mascarton@wpmail.code3.com
Organization: 3M Health Information Systems
X-Newsreader: Forte Agent .99e/32.227
References: <4rqtpi$rjl@nuscc.nus.sg> <4s8jhl$6al@sifon.cc.mcgill.ca> <4t3181$pr9@mack.rt66.com> <4ti605$r00@multivac.orthanc.com>
Date: Mon, 29 Jul 1996 15:53:58 GMT
Lines: 19

lyndon@orthanc.com (Lyndon Nerenberg) wrote:
> Take a look *where* ???  
> --lyndon

Kosko's book on using fuzzy with signal processing applications spends a great
deal of time looking at how to extract and validate the rules from a neural
network.  It's not an area that I've actually built systems to do, so I can't
really comment on it other than to say that it was an excellent read and that it
made intuitive sense (from a set, logic, and programming viewpoint).

The title of the book is "Neural Nets and Fuzzy Systems", 1990, Prentice-Hall.
IMHO an all-around excellent book for folks that are including fuzzy constructs
into existing complex systems.  I was able to use many of the constructs
directly in building a passive sensor (radar) system.

Mark A. Scarton, ABD
CompUtah!, Park City, Utah USA
Home: 801.565.9835
Office: 801.265-4612
