Newsgroups: comp.ai
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!newsfeed.pitt.edu!gatech!news.mathworks.com!tank.news.pipex.net!pipex!dispatch.news.demon.net!demon!sunsite.doc.ic.ac.uk!yama.mcc.ac.uk!news.salford.ac.uk!aber!bath.ac.uk!bsmail!zeus!entpm
From: entpm@zeus.bris.ac.uk (Trevor Martin)
Subject: Re: "Genesis of Fuzzy Concepts", is this a proper research topic?
Message-ID: <Dp6r1v.Aqv@uns.bris.ac.uk>
Sender: usenet@uns.bris.ac.uk (Usenet news owner)
Nntp-Posting-Host: zeus.bris.ac.uk
Organization: University of Bristol, England
X-Newsreader: TIN [version 1.2 PL2]
References: <4jbkd3$csp@infoserv.rug.ac.be>
Date: Mon, 1 Apr 1996 13:34:43 GMT
Lines: 39

Yaxin Liu (yaliu@allserv.rug.ac.be) wrote:
: Since fuzzy concepts are natural according to Lotfi Zadeh and his followers,
: if we can find the genesis of them in human's mind?  

: Obviously this problem relates to the intension and extension of a concept,
: but how a concept will be recognised as a fuzzy one, and how it's 
: membership function is determined are still dubious to me.

: I'm wondering if this question is a good one, a clear one, or it has been
: thoroughly studied.  Please give conments and advises.

: Thanks in advance.

A rule of thumb is that if you can apply a modifier such as "very" or "quite"
to the concept, then it can be modelled using a fuzzy set. For example,
	John is quite tall, 
	Ann's score is very high
suggest that "tall" and "high" are fuzzy concepts.  On the other hand
	John is 1.8 metres 
admits no uncertainty. There has been some
debate on whether probability or fuzziness is appropriate for modelling
uncertainty -  an approach combining the two is discussed in
 Baldwin, J. F., Martin, T. P. and Pilsworth, B. W.
(1995). FRIL - Fuzzy and Evidential Reasoning in AI, Research Studies
Press (John Wiley).
The voting model discussed there also gives a meaning tohe membership function.

comp.ai.fuzzy has recently featured a debate on fuzzy vs probability,
there have also been special issues of journals devoted to the topic.

Trevor Martin

_______________________________________________________________________________
 Advanced Computing Research Centre	||	Phone: +44 117 928 8200
 Department of Engineering Maths	||	Fax:   +44 117 925 1154
 University of Bristol, BS8 1TR, UK	||	email: trevor.martin@bris.ac.uk
_______________________________________________________________________________


