Initializing the EM algorithm in Gaussian mixture models with an unknown number of components

DOI10.1016/j.csda.2011.11.002zbMath1246.65025OpenAlexW2078124810MaRDI QIDQ434890

Igor Melnykov, Volodymyr Melnykov

Publication date: 16 July 2012

Published in: Computational Statistics and Data Analysis (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.csda.2011.11.002

zbMATH Keywords

convergencenumerical examplescluster analysisexpectation-maximization algorithmtruncated normal distributioninitializationeigenvalue decompositionmultivariate Gaussian mixture models

Mathematics Subject Classification ID

Related Items (23)

Cox proportional hazards models with frailty for negatively correlated employment processes ⋮ Model-based clustering of high-dimensional data: a review ⋮ A multivariate linear regression analysis using finite mixtures of \(t\) distributions ⋮ Finite mixture modeling of Gaussian regression time series with application to dendrochronology ⋮ Practical Initialization of Recursive Mixture-Based Clustering for Non-negative Data ⋮ Using mixtures in seemingly unrelated linear regression models with non-normal errors ⋮ Initializing the EM Algorithm for Univariate Gaussian, Multi-Component, Heteroscedastic Mixture Models by Dynamic Programming Partitions ⋮ Multiscale Gaussian convolution algorithm for estimate of Gaussian mixture model ⋮ An effective strategy for initializing the EM algorithm in finite mixture models ⋮ Improved initialisation of model-based clustering using Gaussian hierarchical partitions ⋮ Parsimonious mixtures for the analysis of tensor-variate data ⋮ Flexible mixture modelling using the multivariate skew-\(t\)-normal distribution ⋮ Initialization of Hidden Markov and Semi‐Markov Models: A Critical Evaluation of Several Strategies ⋮ Extending mixtures of factor models using the restricted multivariate skew-normal distribution ⋮ Estimation of finite mixture models of skew-symmetric circular distributions ⋮ Asymmetric clusters and outliers: mixtures of multivariate contaminated shifted asymmetric Laplace distributions ⋮ An extension of the \(K\)-means algorithm to clustering skewed data ⋮ Bayesian approach for mixture models with grouped data ⋮ Adaptive fault detection and diagnosis using parsimonious Gaussian mixture models trained with distributed computing techniques ⋮ Manly transformation in finite mixture modeling ⋮ Using conditional independence for parsimonious model-based Gaussian clustering ⋮ Studying Complexity of Model-based Clustering ⋮ Mixtures of factor analyzers with covariates for modeling multiply censored dependent variables

Uses Software

• mclust
• MixSim
• CARP

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Model-based classification via mixtures of multivariate \(t\)-distributions
• Choosing starting values for the EM algorithm for getting the highest likelihood in multivariate Gaussian mixture models
• Choosing initial values for the EM algorithm for finite mixtures
• Two-way Poisson mixture models for simultaneous document classification and word clustering
• Finite mixture models and model-based clustering
• Variational approximations in Bayesian model selection for finite mixture distributions
• High-dimensional data clustering
• Estimating the dimension of a model
• Maximum likelihood estimation for multivariate skew normal mixture models
• Finding Groups in Data
• Discrete Parameter Variation: Efficient Estimation of a Switching Regression Model
• Algorithms for Model-Based Gaussian Hierarchical Clustering
• A Statistical Model for Positron Emission Tomography
• Convergence properties of a general algorithm for calculating variational Bayesian estimates for a normal mixture model

This page was built for publication: Initializing the EM algorithm in Gaussian mixture models with an unknown number of components