home bbs files messages ]

Forums before death by AOL, social media and spammers... "We can't have nice things"

   comp.ai      Awaiting the gospel from Sarah Connor      1,954 messages   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]

   Message 464 of 1,954   
   j to Cyrus Lan   
   Re: Hidden Markov Modelling problem   
   13 Oct 04 01:26:29   
   
   From: jmontgomery@ea.com   
      
   kclan@hotmail.com (Cyrus Lan) wrote in message news:<416ab62d$1@   
   ews.unimelb.edu.au>...   
   > If the model is like this:   
   >   
   > u_1 -> u_2 -> ... -> u_n   
   > ^       ^             ^   
   > |       |             |   
   > d_1 -> d_2 -> ... -> d_n   
   >   
   > then the model using first order will become:   
   > p(u, d) = sigma p(u_i|d_i)p(d_i|d_i-1)   
   >   
   > Now there are relations r imposed between each pair of u usch that:   
   >   
   >   r_1         r_i   
   >    |        |    |   
   >  |    |   |        |   
      
   > |       |            |   
   > u_1 -> u_2 -> ... -> u_n   
   > ^       ^             ^   
   > |       |             |   
   > d_1 -> d_2 -> ... -> d_n   
   >   
   > how can I formulate the model? Thankyou very much.   
   >   
      
   The actual joint is given by:   
      
     Pr(ui,di) = Pr(ui|ui-1,di)*Pr(di|di-1)*Pr(ui-1,di-1)   
      
   A relation between ui and ui-1 can be encoded in Pr(ui|ui-1,di).   
      
   j   
      
   [ 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)   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]


(c) 1994,  bbs@darkrealms.ca