From: nonlinear@rogers.com   
      
   Dear Jose   
      
   It could be that the cluster at either end of the data y is such that it   
   covers the extreme range-that is, it has the >>reach<< to do so!!.   
      
   In principle, I do not think this is a contradiction.   
      
   Paul   
      
   José Victor wrote:   
   > Fuzzy clustering algorithms are frequently used to determine the structure   
   > of fuzzy models (rules). Most of the fuzzy clustering algorithms make use of   
   > a distance measure to partitioning the data and there is a concern to   
   > guarantee more or less a uniform distribution of the cluster centres over   
   > the data.   
   > I wonder if this is the best strategy. I think it was Kosko who proved that   
   > "optimal fuzzy rules cover extrema", so isn't here some contradiction?   
   >   
   > JV   
   >   
   >   
   >   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)