XPost: comp.ai.neural-nets, sci.image.processing, sci.math.num-analysis   
      
   Milind wrote:   
   > Hi Peter,   
   >   
   > I agree with Greg and Ted. Try no algorithm before understanding your   
   > data really well.   
      
   Hi Milind,   
      
   Many thanks for your reply.  I have a correlation matrix showing the   
   correlations of the metrics with one another and with the errors that I   
   am hoping to use to classify the observations.  One class is for errors   
   below a threhold and the other class is for errors above that threhold.   
    The metrics are attributes that I use to try to predict whether the   
   unknown errors are above or below the threshold.  I was thinking of   
   selecting the metrics that have the highest correlation with the errors   
   and the lowest correlation with one another.  However, it appears that   
   I could go straight to what I want with principal component analysis.   
   Do you agree?   
      
   Thanks again,   
   Peter.   
      
   [ 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. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)