Hidden Markov Models with mixtures as emission distributions

被引:28
|
作者
Volant, Stevenn [1 ,2 ]
Berard, Caroline [1 ,2 ]
Martin-Magniette, Marie-Laure [1 ,2 ,3 ,4 ,5 ]
Robin, Stephane [1 ,2 ]
机构
[1] INRA, UMR MIA 518, F-75231 Paris, France
[2] AgroParisTech, UMR MIA, F-75231 Paris, France
[3] INRA, URGV UMR1165, F-91057 Evry, France
[4] UEVE, UMR URGV, F-91057 Evry, France
[5] CNRS, UMR URGV ERL8196, F-91057 Evry, France
关键词
Hidden Markov models; Model-based clustering; Mixture model; Hierarchical algorithm;
D O I
10.1007/s11222-013-9383-7
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
In unsupervised classification, Hidden Markov Models (HMM) are used to account for a neighborhood structure between observations. The emission distributions are often supposed to belong to some parametric family. In this paper, a semiparametric model where the emission distributions are a mixture of parametric distributions is proposed to get a higher flexibility. We show that the standard EM algorithm can be adapted to infer the model parameters. For the initialization step, starting from a large number of components, a hierarchical method to combine them into the hidden states is proposed. Three likelihood-based criteria to select the components to be combined are discussed. To estimate the number of hidden states, BIC-like criteria are derived. A simulation study is carried out both to determine the best combination between the combining criteria and the model selection criteria and to evaluate the accuracy of classification. The proposed method is also illustrated using a biological dataset from the model plant Arabidopsis thaliana. A R package HMMmix is freely available on the CRAN.
引用
收藏
页码:493 / 504
页数:12
相关论文
共 50 条
  • [21] Hidden Markov models
    Eddy, SR
    CURRENT OPINION IN STRUCTURAL BIOLOGY, 1996, 6 (03) : 361 - 365
  • [22] Feature Selection for Hidden Markov Models and Hidden Semi-Markov Models
    Adams, Stephen
    Beling, Peter A.
    Cogill, Randy
    IEEE ACCESS, 2016, 4 : 1642 - 1657
  • [23] On mixtures of distributions of Markov chains
    Fortini, S
    Ladelli, L
    Petris, G
    Regazzini, E
    STOCHASTIC PROCESSES AND THEIR APPLICATIONS, 2002, 100 : 147 - 165
  • [24] Markov models — training and evaluation of hidden Markov models
    Jasleen K. Grewal
    Martin Krzywinski
    Naomi Altman
    Nature Methods, 2020, 17 : 121 - 122
  • [25] Markov models - training and evaluation of hidden Markov models
    Grewal, Jasleen K.
    Krzywinski, Martin
    Altman, Naomi
    NATURE METHODS, 2020, 17 (02) : 121 - 122
  • [26] Approximate posterior distributions for convolutional two-level hidden Markov models
    Rimstad, Kjartan
    Omre, Henning
    COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2013, 58 : 187 - 200
  • [27] Consistent Estimation of the Filtering and Marginal Smoothing Distributions in Nonparametric Hidden Markov Models
    De Castro, Yohann
    Gassiat, Elisabeth
    Le Corff, Sylvain
    IEEE TRANSACTIONS ON INFORMATION THEORY, 2017, 63 (08) : 4758 - 4777
  • [28] Finite Mixtures of Hidden Markov Models for Longitudinal Responses Subject to Drop out
    Marino, Maria Francesca
    Alfo, Marco
    MULTIVARIATE BEHAVIORAL RESEARCH, 2020, 55 (05) : 647 - 663
  • [29] Quantifying copy number variations using a hidden Markov model with inhomogeneous emission distributions
    Mccallum, Kenneth Jordan
    Wang, Ji-Ping
    BIOSTATISTICS, 2013, 14 (03) : 600 - 611
  • [30] USE OF DIFFERENT NUMBERS OF MIXTURES IN CONTINUOUS DENSITY HIDDEN MARKOV-MODELS
    CHUNG, YJ
    UN, CK
    ELECTRONICS LETTERS, 1993, 29 (09) : 824 - 825