From: rif@mit.edu   
      
   Certainly it depends heavily on the classifier.  I have successfully   
   trained linear SVMs with orders of magnitude more features than   
   examples.  For instance,   
      
   Ramaswamy, Tamayo, Rifkin, Mukherjee, Yeang, Angelo, Ladd, Reich,   
   Latulippe, Mesirov, Poggio, Gerlad, Loda, Lander and   
   Golub. "Multiclass cancer diagnosis using tumor gene expression   
   signatures." Proceedings of the National Academy of Science, vol. 98,   
   no. 26, 18 December 2001.   
      
   Cheers,   
      
   rif   
      
   [ comp.ai is moderated ... your article may take a while to appear. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)