From: jakulin@@ieee.org   
      
   > I am starting a Ph.D. about  Building Intellegent Expert system for   
   > diagnosis of medical images. I know there are two schools for doing   
   > that: the numeric school and the symbolic school.   
      
   You have to ask yourself whether you will be a) putting human   
   knowledge into the system (symbolic approaches are more appropriate),   
   or b) acquiring knowledge inductively from symptoms (symbolic   
   approaches are better) or from images (numerical approaches are less   
   constrained). Whatever you do, you better make sure that you properly   
   deal with uncertainty: you need to deal with statistics, either   
   Fisherian or Bayesian. Once you deal with uncertainty, there is no   
   major distinction between symbolic and numeric approaches. Bayesian   
   approaches have a neat way of including human knowledge into the model   
   in the form of priors.   
      
   There are many journals and conferences. Ask your thesis advisor to   
   give you directions and tell you which conferences he attends and   
   which journals he reads. You should start your research in the field   
   of your advisor, otherwise you may get lost.   
      
   Aleks   
      
   [ 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)