2
$\begingroup$

I'm working with Hidden Markov Models and I have a dataset composed by independent phrases, where each word is an observation. Hence, the best way to adjust my parameters (via Baum-Welch algorithm) is considering each phrase per time and not all phrases concatenated.

I would like to know if there is an algorithm that do the training in this way. If not, what are the strategy to avoid transitions created by the concatenation (last word of to first word).

| cite | improve this question | | | | |
$\endgroup$
1
$\begingroup$

In that case, you would typically add a special "start" and "stop" symbol to the beginning and end of each sequence, and then concatenate them. Then define a special start and stop state which only outputs its special symbol.

| cite | improve this answer | | | | |
$\endgroup$
  • $\begingroup$ Do you know if there is a method for online training? $\endgroup$ – zeferino Mar 15 '13 at 21:50
  • $\begingroup$ I'm only familiar with online techniques used with Bayesian HMMs such as particle filtering, of which there are many variations. But I'm sure there are non-Bayesian methods as well. $\endgroup$ – jerad Mar 15 '13 at 21:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.