Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!samsung!noose.ecn.purdue.edu!lips.ecn.purdue.edu!kavuri
From: kavuri@lips.ecn.purdue.edu (Surya N Kavuri )
Newsgroups: comp.ai.neural-nets
Subject: Radial-basis-functions and KMeans
Message-ID: <1991May19.231036.2422@noose.ecn.purdue.edu>
Date: 19 May 91 23:10:36 GMT
Sender: root@noose.ecn.purdue.edu (ECN System Management)
Organization: Purdue University Engineering Computer Network
Lines: 14


 
 I have a question regarding the use of Kmeans clustering to 
 find the centers and radii of radial basis functions ?

 How does this algorithm work when:
 (a) classes overlap
 (b) classes are adjacent          

 Moody and Darken (1989 Neural Comp.) do not show how their method
 handles these cases

 SURYA  kAVURI
 (FIAT LUX)