From: tdunning@san.rr.com   
      
   Recurrent neural nets are probably massive overkill for a problem like   
   this.   
      
   You should start with sub-string presence or absence as a proxy for   
   string edit distance.  You can even use the sub-strings to generate   
   hints for dynamic programming based edit distance systems.   
      
   Remember that weighted sub-string matching is just a handy way of   
   encoding a Markov model.  As such, it has enormous power for doing   
   string matching sorts of operations.   
   What made you think that fancy stuff was needed?   
      
   [ comp.ai is moderated.  To submit, just post and be patient, or if ]   
   [ that fails mail your article to , and ]   
   [ ask your news administrator to fix the problems with your system. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)