Newsgroups: comp.ai.fuzzy
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!ix.netcom.com!netcom.com!dfuess
From: dfuess@netcom.com (David A. Fuess)
Subject: Re: Membership functions and non-existant data.
Content-Type: text/plain; charset=us-ascii
Message-ID: <dfuessD7FvB0.9wp@netcom.com>
Sender: dfuess@netcom2.netcom.com
Content-Transfer-Encoding: 7bit
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
X-Newsreader: QNews v0.9b3 Beta 14 Apr 1994 Evaluation copy.
References: <D7Bv35.DBn@news.hawaii.edu> <3n95dv$l29@nntp.crl.com>
Mime-Version: 1.0
Date: Sat, 22 Apr 1995 15:38:18 GMT
Lines: 20

In article <3n95dv$l29@nntp.crl.com>
Richard Berman <sts@crl.com> wrote:
> jamesw@uhunix.uhcc.Hawaii.Edu (James Williams) wrote:
[snip]
> > value of 0 for x=0.  But if you don't have a value for x, then
> > what is the value of the membership function?  
[snip]
>  The membership value measures compatibility.  The value submitted
>  is outside the domain, and so is incompatible.

I disagree. If the fuzzy set implements an open ended set for the notion of 
large in some universe of discourse, then any value greater than the endpoint 
would be large also. Perhaps the definition of the universe of discourse for 
the set in question is flawed, obviously at least one value lies outside the 
definition boundries for the set. The set should be expanded as it does not 
yet cover the true universe of discourse.

Another way to handle the problem is to write open ended quantizers. (e.g.
if (V > MAXV) then V=MAXV), hence, any valid input is converted to the range 
of the set.
