Make Entry -- Become an Editor -- Most Popular: (10, 25, 50, 75, 100)

   Help

Hidden Markov Model

A Hidden Markov Model or HMM is a type of statistical model in which the system that is being modeled is marked as a Markov process that do not have any known parameters, and the goal or objective is to find out the hidden parameters from the observable parameters. The model parameters that are taken out can then be used in order to do further analysis – for example, in such applications as pattern recognition. Many people consider the Hidden Markov Model as the simplest dynamic Bayesian network.

There are certain characteristics that differentiate a Hidden Markov Model from a regular Markov Model. In a regular Markov Model, the state is directly visible to the observer and because of this the only parameters used are the state transition probabilities. On the other hand, a Hidden Markov Model does not have a directly visible state, what is visible though are the variables that have been influenced by the state. Each particular state has a probability distribution over the possible output tokens. What this means therefore is that the sequence of tokens generated by a Hidden Markov Model provides information about the sequence of states.

Web Resources On Hidden Markov Model

Hidden Markov Models
Definition of Hidden Markov Model

Book Resources On Hidden Markov Model

Inference in Hidden Markov Models by Olivier CappÈ, Eric Moulines, Tobias Ryden
Hidden Markov Models of Bioinformatics by Timo Koski

Site Maps: Most Recent | Clusters | Browse
New: Graduate Student Job Opportunity