From: grs-NO-SPAM@NO-mail.com   
      
   > (...) An advantage of the fuzzy   
   > intuitionistic approach is that "unknown" is as legal as any other value.   
   > So C in fact has a value = "unknown". As some rules get fired they impose   
   > constraints on C, thus making it more and more certain.   
      
   The main advantage of such "intermediate" fuzzy variables is to allow   
   the expression of, as I said before, "somewhat" abstract concepts.   
   It's roughly the equivalent of multiple hidden layers in neural   
   networks. With the advantage of fuzzy logic, which is, amongst   
   other things, a strongly semantics-oriented system. To put it simply,   
   fuzzy rules "make sense" by themselves, whereas it's extremly   
   difficult to say what is doing what in a moderately complex   
   neural network-based system.   
      
   I'll take an example: let's suppose a vision-oriented fuzzy system   
   with a certain number of inputs connected to a set of vision "sensors"   
   (I'm not getting into great details).   
   Now, you could elaborate several sets of fuzzy rules, with sequential   
   dependencies, to express more an more abstract vision concepts: for   
   instance, a preliminary set of rules could transform input pixels   
   to very basic shapes like straight lines, high contrast edges, etc.   
   Then additional intermediate rules could transform the basic concepts   
   to more abstract concepts, like "complex" geometrical shapes, and could   
   finally lead to character or shape recognition.   
      
   > I think that if implemented properly it cannot introduce any additional   
   > instability.   
      
   Possibly, but by manipulating "fuzzy abstract concepts" in the way I   
   stated above, I think instability is more likely to happen.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)