From: project2501@project2501.cor   
      
   i'm looking into a clustering algorithm (modififcation of standard k-means   
   to start with) for fuzzy data.   
      
   a simplified 3-dimensional data instance may take the form:   
   	(   
   	length = short/0.2 + medium/0.8 + long/0.3, texture = smooth/0.3 +   
   	rough/0.7,   
   	colour = red/0.3 + blue/0.7 + green/0.1 )   
      
   where short, medium, long are predefined (different) shapes, or fuzzy sets   
   on length. and similar for texture and colour.   
      
   my question is this: are there existing standard distance measures between   
   data instance of this type? i have spent some time looking around the web,   
   citeseer, etc ... and have reach the opinion that there is no generally   
   agreed upon general measure? (keeping the scaling ussue aside, assume   
   normalised according to application).   
      
   even if we use the standard euclidian distances, how does one combine the   
   components:   
   	(length1^2 + length2^2) + (texture1^2 + texture2^2) + ... ?   
      
      
   thanks for opinions, advice or pointers.   
      
   p   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)