From: info@fretfarm.co.uk   
      
   All,   
      
   Sorry if this is a common question, couldn't find it on the FAQ.   
      
   In fuzzy control we map inputs to a set of fuzzy states, e.g.:   
      
   a -> SMALL, MEDIUM, LARGE   
   b -> HOT, MILD   
      
   where each set spans all inputs (i.e. degree of membership sums to one for   
   states belonging to one input)   
      
   Then rules consist conjunctions of all combinations of these states (1 per   
   input) with a corresponding output state, e.g.:   
      
   if SMALL and HOT then CHILLE   
   if SMALL and MILD then PEPPER   
   if MEDIUM and HOT then CHILLE   
   if MEDIUM and MILD then PEPPER   
   if LARGE and HOT then PEPPER   
   if LARGE and MILD then PEPPER   
      
   I came across the posit:   
      
   any set of (consistent) rules relating input states to output states, using   
   any operator (AND, OR, NOT, XOR, etc) can be transformed into a rule-set of   
   the above type.   
      
   For example, say we have the rules:   
   if LARGE or MILD then PEPPER   
   if not LARGE and not MILD then CHILLE   
      
   These are directly transformed into the rule-block above.   
      
   The person I was talking to said that this is the _whole point_ of having   
   such rule blocks (with all combinations). But I couldn't google any support   
   for his theory.   
      
   Is this true? and if so, is there a general procedure to transform the   
   arbitrary rules into block form?   
      
      
   Thanks in advance   
      
   Ian.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)