search for




 

Classification analysis using hidden Markov model
Journal of the Korean Data & Information Science Society 2021;32:1135-41
Published online September 30, 2021;  https://doi.org/10.7465/jkdi.2021.32.5.1135
© 2021 Korean Data and Information Science Society.

Yeji Cheon1 · Hayoung Choi2 · Yongku Kim3

1Korea Real Estate Board
23Department of Statistics, Kyungpook National University
Correspondence to: 1 Assistant manager, Korea Real Estate Board, Daegu 41068, Korea.
2 Graduate student, Department of Statistics, Kyungpook National University, Daegu 41566, Korea
3 Corresponding author: Associate professor, Department of Statistics, Kyungpook National University, Daegu 41566, Korea. E-mail: kim.1252@knu.ac.kr
This researchis based on the part of Yeji Cheon's Master thesis and was supported by the Research Grants of Korea Forest Service (Korea Forestry Promotion Institute) project (No.2019149A00-2123-0301).
Received July 30, 2021; Revised August 15, 2021; Accepted August 17, 2021.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
A hidden Markov model (HMM) provides useful representations of dependent heterogeneous phenomena. So it becomes a popular method for modelling stochastic pro-cesses and time-dependent sequences, and is primarily applied in many different fields such as language, handwriting recognition, and molecular biology. Especially, in the sequence classification case, classification among known hidden Markov models is known to be accomplished with a classifier that minimizes the probability of error. In this pa- per, we first generate variables for the hidden state using the hidden markov model and then analyze the state using various classification methods. It differs from the existing analysis method by using the state variable and the mixture distribution based on the state rather than using the observed value directly in the analysis. In addition, it can be used to identify the relevance in the underlying process. As as illustration, we used the annual production of Matsutake mushroom data observed in five regions from 1997 to 2016.
Keywords : Classification analysis, hidden Markov moddel, Matsutake mushroom, time-dependent process.