From: LNMEgo@hotmail.com   
      
   I think I understand bayasian theory ok.  I understand that if I   
   choose a discreet value for a node then it chages the probabilities of   
   the values in other nodes.  I am working on something that is similar   
   to spam filtering, except that the data I'm looking to predict won't   
   be binary (spam vs no spam).  I can't just give particular words a   
   high probability of predicting "true".  For example, I need "chest"   
   and "pain" to predict "chest pain".  I have made networks like this   
   before, the problem is that the nature of the data I am supplying is   
   different.  A list of words (which is what makes it similar to spam   
   detection) is different from an organized table of data.   
      
   So, how do I set up and train a network to do this.  Do I set up a   
   node for each word in the string (1st word, 2nd word, etc) and train   
   it on all permutations of the word list? (I think this might over   
   train it to the longer ones.) Do I not understand something about the   
   theory?   
      
   Thanks in advance.   
      
   [ 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)