From: fbenavent@cogitoergosum.com   
      
   Hi Ryan,   
      
   > If I have a 100 feature sets, (each containing say 7 items), how does   
   > this get mapped to a two dimensional graph to visualize a cluster? I   
   > could understand if my feature set contained two items (map each item   
   > to an axis).   
      
   What you are looking for is caled MDS (Multidimensional scaling), is   
   an algorithm to obtain a set of bidimensional points which had the   
   property that distances between each two points are the same that the   
   "distance" between the original n-dimensional sets.   
      
   The 2D points that you obtain are only usefull as a visualization   
   tool, but sometimes it is needed. Some time ago I found some articles   
   about it but I dont remember when, try to find:   
      
   - "An Introduction to MDS" by Florian Wickelmaier   
      
   - "Multidimensional Scaling, Encyclopedia of Cognitive Science" by   
   Matk Steyvers   
      
   And a PowerPoint tutorial called "Lecture 5: Multidimensional Scaling   
   and Cluster Analysis", sorry no authorship available.   
      
   regards,   
      
   Francesc   
      
   [ comp.ai is moderated.  To submit, just post and be patient, or if ]   
   [ that fails mail your article to , and ]   
   [ ask your news administrator to fix the problems with your system. ]   
   ------------ And now a word from our sponsor ------------------   
   For a quality usenet news server, try DNEWS, easy to install,   
   fast, efficient and reliable. For home servers or carrier class   
   installations with millions of users it will allow you to grow!   
   ----  See http://netwinsite.com/sponsor/sponsor_dnews.htm  ----   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)