Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!newsfeed.pitt.edu!godot.cc.duq.edu!news.duke.edu!news.mathworks.com!newsfeed.internetmci.com!in1.uu.net!news.interpath.net!sas!newshost.unx.sas.com!saswss
From: saswss@hotellng.unx.sas.com (Warren Sarle)
Subject: Re: How to classify with only 1 class?
Originator: saswss@hotellng.unx.sas.com
Sender: news@unx.sas.com (Noter of Newsworthy Events)
Message-ID: <DquvoC.6GD@unx.sas.com>
Date: Sat, 4 May 1996 00:50:36 GMT
X-Nntp-Posting-Host: hotellng.unx.sas.com
References: <4lhhvr$693@nntp5.u.washington.edu> <4lhvip$8pe@ss10.elvis.ru> <4ljgsm$cp9@news.sandia.gov> <4lpg63$7gi@dfw-ixnews8.ix.netcom.com> <831037859.19991.0@intechfs.demon.co.uk>
Organization: SAS Institute Inc.
Lines: 26


In article <831037859.19991.0@intechfs.demon.co.uk>, Gavin Smith <nrl@intechfs.demon.co.uk> writes:
|> jdadson@ix.netcom.com(Jive Dadson ) wrote:
|> >...
|> >What you could do is use a density estimator on the one-class training
|> >set. Then if your unclassified datum came from a low-density area you
|> >could speculate that the datum MIGHT not be from the class, but there
|> >would be no way to put a confidence bound on that without at least
|> >guessing about the density distribution of "others".
|> >
|> One approach is to use Unsupervised Hebbian Learning with only one output 
|> unit.

Hebbian learning with one output amounts to computing the first
principal component, so this approach would work only if the single
class is concentrated near a straight line in the input space.

Jive's suggestion is much more useful.



-- 

Warren S. Sarle       SAS Institute Inc.   The opinions expressed here
saswss@unx.sas.com    SAS Campus Drive     are mine and not necessarily
(919) 677-8000        Cary, NC 27513, USA  those of SAS Institute.
