PERFORMANCE IMPROVEMENT OF THE ESTIMATION ALGORITHM FOR AN OBSERVED SEQUENCE IN HIDDEN MARKOV MODELS ON THE BASIS OF ALGEBRAIC BAYESIAN NETWORKS

Mikhail Pinsky, Sirotkin Alexander Vladimirovich, Andrey A. Filchenkov

2011 , VOLUME 11, NUMBER 5 ( SEPTEMBER-OCTOBER )

ISSN 2226-1494 (print), ISSN 2500-0373 (online)

Publications

Editor-in-Chief

Nikiforov
Vladimir O.
D.Sc., Prof.

Partners

PERFORMANCE IMPROVEMENT OF THE ESTIMATION ALGORITHM FOR AN OBSERVED SEQUENCE IN HIDDEN MARKOV MODELS ON THE BASIS OF ALGEBRAIC BAYESIAN NETWORKS

M. Pinsky, A. V. Sirotkin, A. A. Filchenkov

Read the full article

Abstract

Hidden Markov models and algebraic Bayesian networks are probabilistic graphical models and therefore they are quite similar. Hidden Markov models have wide application while algebraic Bayesian networks are not so widespread, but their instruments allow simulating and solving hidden Markov models problems. We considered the improvement performance of the solution algorithm for hidden Markov models first problem by means of algebraic Bayesian networks methods. An algorithm for estimating probability of observed sequence in binary linear hidden Markov models by means of algebraic Bayesian network posterior inference is formulated in the article.

Keywords: hidden Markov models, algebraic Bayesian networks, binary linear hidden Markov models, probabilistic graphical models.

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License