XPost: comp.ai.edu   
   From: albinali@cs.arizona.edu   
      
   Andrew Guillory  wrote in message news:   
   41ae6e38$1@news.unimelb.edu.au>...   
   > Fahd wrote:   
   > > I have binary vectors that represent sensor readings where each entry can   
   > > take a value of 0 or 1 (Active sensor/ inactive sensor). I am trying to use   
   > > this data to train a Discrete Hidden Markov Model which requires   
   > > transforming the data vectors to discrete observations. Any suggestions on   
   > > how to do that? Thanks.   
   >   
   > Have you tried just treating the binary vector as a binary integer?   
   >   
   > int number = sensor1 + 2 * sensor2 + 4 * sensor3 + 8 * sensor4...   
   >   
   > Maybe there is a reason for not doing this that I missed?   
   >   
   > Andrew Guillory   
   >   
      
   Well. Here is the problem, there might be different binary vectors for   
   the same discrete observation and therefore I suppose I would need a   
   more elaborate method. Any thoughts on how to do that with multiple   
   vectors? Thanks in advance!   
      
   Fahd   
      
   [ 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)