77064bd3   
   From: mailbox@dmitry-kazakov.de   
      
   On Thu, 17 Sep 2009 14:17:13 -0700 (PDT), Nate wrote:   
      
   > The following fuzzy algorithm showed up in a paper I'm trying to   
   > understand:   
   >   
   > Fuzzy rule 1: if ek is LN then ek' is LN (highest speed).   
   > Fuzzy rule 2: i f ek is SN then ek' is SN (high speed).   
   > Fuzzy rule 3: if ek is ZE then ek' is ZE (normal speed).   
   > Fuzzy rule 4: if ek i s SP then ek' is SP (slow speed).   
   > Fuzzy rule 5: if ek is LP then ek' is LP (lowest speed).   
   >   
   > In the fuzzy algorithm, a linguistic rule is used to calculate the   
   > modified error ek'.  ek is the predicted error.   
   >   
   > What could they mean by ek' is LN.   
      
   ek' = LN ?   
      
   > Don't you fuzzify, find   
   > membership, and get a crisp value which you call ek'? What does it   
   > matter that ek' is LN?   
      
   It does matter if ek' is defined over another domain than ek, like over the   
   set of linguistic variables:   
      
      ek' : { LN, SN, ZE, SP, LP } -> [0,1] , a fuzzy subset of {LN, SN.. LP}   
      
   While ek is defined over the domain of velocities:   
      
      ek : ] -oo m/s, +oo m/s [ -> [0,1],  a fuzzy subset of velocities   
      
   Or maybe ek is crisp velocity:   
      
      ek in ] -oo m/s, +oo m/s [   
      
   then the above is just fuzzification of speed into a fuzzy subset of   
   linguistic variables {LN, SN ... LP}. I.e. merely   
      
      ek' = fuzzified ek   
      
   (In which case I would consider it wasting time to read papers deploying a   
   notation like above instead of saying that ek' is a fuzzified speed! (:-()   
      
   --   
   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)