Markov models for linguistic sequences

Authors

  • Ihor Vsevolodovych Baklan
  • Tetiana Viktorivna Shulkevych

DOI:

https://doi.org/10.34185/1562-9945-4-123-2019-09

Keywords:

прихована Марковська модель, лінгвістична модель, лінгвістичне моделювання

Abstract

The popularity of hidden Markov models of fuels and lubricants and their implementation in various fields, spreads every year, leads to certain problems.
The aim of the study is the use of hidden Markov models for the analysis of time series in the form of linguistic chains.
This study has its own goal of identifying the problems facing the developers of intelligent systems with the use of fuel and lubricants and identifying some of the areas in which these problems can be overcome. For the whole family of standard fuels and lubricants, three main problems were identified, the solution of which is very important for analyzing and forecasting time series.
Today, hidden Markov models is one of the most common mathematical tools used for many classifiers and modeling of various problems. In recent years, fuels and lubricants are used for gesture recognition. It is clear that this article does not provide a complete list of the problems facing the developers of intelligent systems using fuel, but it is a definite step towards the integration of modern methods for solving complex problems.

References

Timo Koski. Hidden Markov Models for Bioinformatics. – Dordrecht: Kluwer Academic Publicher, 2001. – 392 p.

S.E.Levinson, L.R.Rabiner and M.M.Sondhi An Introduction to the Applications of Theory of Probabilistic Functions of a Markov Chain to Automatic Speech Recognition. – The Bell System Technical Journal, 1983, 62, pp.1053-1074.

Z.Grahramani and M.Jordan. Factorial hidden Markov models. – Machine Learning, 1997, 29, pp.245-273.

R.J.Boys, D.Henderson and D.J.Wilkinson. Detecting homogeneous segments in DNA sequences using hidden Markov models. Applied Stistacs, 2000, 49, Part 2, pp.269-285.

Y.Ephraim, A.Dembo and L.R.Rabiner. Minimum Discrimination Information Approach for Hidden Markov Modelling. – IEEE Trans. jn Information Theory, 1989, 35, pp.1000-1013.

P.Baldi and Y.Chauvin. Smooth On-Line Learning Algorithms for Hidden Markov Models. – Neural Computation, 2000, 6, pp. 307-318.

B.H. Juang and L.R. Rabiner. The segmental K-means Algorithm for Estimating Parameters of Hidden Markov Models. IEEE Transactions on Acoustics, Speech and Signal Processing, 1990, 38, pp. 1639-1641.

H. Lucke. Which Stochastic Models Allow Baum-Welch Training? - IEEE Trans¬actions on Signal Processing, 1996, 44, pp. 2746—2756.

P. Smyth, D. Heckerman and M.L. Jordan. Probabilistic Independence Networks for hidden Markov probability models. Neural Computation,1997, 9, pp. 227—269.

J.Q.Smith. Influence diagramsfor statistical modelling. Annals of Statistics, 1989, 17, pp.654-672.

Y. Shen, E. Muth, and A. Hoover Senior Member The Impact of Quantity of Training Data on Recognition of Eating Gestures / Preprint arXiv:1812.04513v1 [cs.LG] 11 Dec 2018

Baklan I.V., Stepankova G.A. KlasifIkatsIya modeley markovskogo tipu: naukova monografIya . - K.: NAU, 2012. – 84 s.

Baklan I., Komada P. Hybrid hidden Markov models // Elektronika (LIV). - No 8/2013. – P.28-31.

Published

2019-10-12