From: mailbox@dmitry-kazakov.de   
      
   On 2 Feb 2005 21:05:55 -0800, Balaji wrote:   
      
   > I am pursuing my Master's degree in computer science. As a part of my   
   > curriculam, i have undertaken a project to build a fuzzy inference   
   > shell. I wish to incorporate weights for rules and the input domains in   
   > the proposed shell. Can some one help me with literature for weighted   
   > Compositional rule of inference and Weighted compatability modification   
   > inference   
      
   In my view there should be no place for any weights in a truly fuzzy   
   inference system. A rule may have either a truth level or   
   possibility/necessity describing our belief in this rule in *fuzzy* terms,   
   i.e. within the same fuzzy framework. But if it had a weight, that would   
   assume existence of some parallel, competing model of uncertainty/belief.   
   If this model is better than fuzzy, then use it instead. If it is worse,   
   then throw it away!   
      
   Of course we could imagine cases like randomly selected fuzzy rules, where   
   inherence would be fuzzy and weights would be the probabilities that a rule   
   gets selected, but such system as a whole won't be fuzzy. That will be   
   random fuzzy sets.   
      
   --   
   Regards,   
   Dmitry A. Kazakov   
   http://www.dmitry-kazakov.de   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)