Newsgroups: comp.ai.fuzzy,sci.stat.math
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!news.sprintlink.net!unify!sxl
From: sxl@unify.com (Steve Lembark)
Subject: Re: Fuzzy theory or probability theory?
Message-ID: <D040EE.CpI@unify.com>
Sender: usenet@unify.com
Nntp-Posting-Host: fantasy
Organization: Unify Corporation, Sacramento, California
References: <3af1jc$3fg@nuscc.nus.sg> <3aqfdr$11b1@hearst.cac.psu.edu> <Czq78n.8vG@news.cis.umn.edu> <3bdbgl$15jm@hearst.cac.psu.edu>
Date: Thu, 1 Dec 1994 02:03:50 GMT
Lines: 20
Xref: glinda.oz.cs.cmu.edu comp.ai.fuzzy:3501 sci.stat.math:3408

Fuzzy logic is a synthesis.  As such its margins are... well... fuzzy.
Saying that fuzzy logic is "just probability" or "just logic" is like
saying that chemistry is "just physics at STP".  Chemists use physcal
laws but go well beyond what physics alone can offer.

The same holds true for fuzzy logic.  It draws heavily on both traditional
boolean logic and probability theory -- lacking either it couldn't exist.
There are a range of problems which neither alone has been able to successfully
solve however.  Fuzzy fills in the gap.

Another way to say it might be that if either traditional logic or
probablity theory had been able to solve these problems by themselves
noone would have bothered to create fuzzy logic.  As it stands they can't
and the new synthesis was developed to handle the areas where each fell
down.  The synthesis has its own advances which go beyond either of its
two base subjects.  None of them is "better", they all work well on the
problems they were developed to solve -- each addressing a different set.

steve lembark
(sxl@unify.com)
