Estimating number of components in Gaussian mixture model using combination of greedy and merging algorithm

Štěpánová, Karla; Vavrecka, Michal

doi:10.1007/s10044-016-0576-5

Cited by 4 publications

(3 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…., K components instead. Subsequently, the optimized greedy initialization methods are proposed [19,20]. Another method is to apply Rough-Enhanced-Bayes mixture estimation (REBMIX) to the initialization of the EM algorithm [21].…”

Section: Plos Onementioning

confidence: 99%

“…In the paper of Verbeek et al [ 18 ], a greedy algorithm is presented which does not need an initialization but needs to construct a whole sequence of mixture models with m = 1, …, K components instead. Subsequently, the optimized greedy initialization methods are proposed [ 19 , 20 ]. Another method is to apply Rough-Enhanced-Bayes mixture estimation (REBMIX) to the initialization of the EM algorithm [ 21 ].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A new iterative initialization of EM algorithm for Gaussian mixture models

You

2023

PLoS ONE

View full text Add to dashboard Cite

Background The expectation maximization (EM) algorithm is a common tool for estimating the parameters of Gaussian mixture models (GMM). However, it is highly sensitive to initial value and easily gets trapped in a local optimum. Method To address these problems, a new iterative method of EM initialization (MRIPEM) is proposed in this paper. It incorporates the ideas of multiple restarts, iterations and clustering. In particular, the mean vector and covariance matrix of sample are calculated as the initial values of the iteration. Then, the optimal feature vector is selected from the candidate feature vectors by the maximum Mahalanobis distance as a new partition vector for clustering. The parameter values are renewed continuously according to the clustering results. Results To verify the applicability of the MRIPEM, we compared it with other two popular initialization methods on simulated and real datasets, respectively. The comparison results of the three stochastic algorithms indicate that MRIPEM algorithm is comparable in relatively high dimensions and high overlaps and significantly better in low dimensions and low overlaps.

show abstract

Section: Plos Onementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A new iterative initialization of EM algorithm for Gaussian mixture models

You

2023

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…Here we used the default which is one initialization which is obtained from a kmeans clustering of the data [19]. Note that there are other methods for fitting a GMM [21][22][23] some of which may be more efficient, less sensitive to initialization, or have additional benefits such as estimating n. The fitting time generally increases as P and n increase. Ref.…”

Section: Gaussian Mixture Modelmentioning

confidence: 99%

The Use of Gaussian Mixture Models with Atmospheric Lagrangian Particle Dispersion Models for Density Estimation and Feature Identification

Crawford

2020

Atmosphere

View full text Add to dashboard Cite

Atmospheric Lagrangian particle dispersion models, LPDM, simulate the dispersion of passive tracers in the atmosphere. At the most basic level, model output consists of the position of computational particles and the amount of mass they represent. In order to obtain concentration values, this information is then converted to a mass distribution via density estimation. To date, density estimation is performed with a nonparametric method so that output consists of gridded concentration data. Here we introduce the use of Gaussian mixture models, GMM, for density estimation. We compare to the histogram or bin counting method for a tracer experiment and simulation of a large volcanic ash cloud. We also demonstrate the use of the mixture model for automatic identification of features in a complex plume such as is produced by a large volcanic eruption. We conclude that use of a mixture model for density estimation and feature identification has potential to be very useful.

show abstract

Finite mixture modeling in time series: A survey of Bayesian filters and fusion approaches

Liang

Xiao

et al. 2023

Information Fusion

View full text Add to dashboard Cite

Estimating number of components in Gaussian mixture model using combination of greedy and merging algorithm

Cited by 4 publications

References 37 publications

A new iterative initialization of EM algorithm for Gaussian mixture models

A new iterative initialization of EM algorithm for Gaussian mixture models

The Use of Gaussian Mixture Models with Atmospheric Lagrangian Particle Dispersion Models for Density Estimation and Feature Identification

Finite mixture modeling in time series: A survey of Bayesian filters and fusion approaches

Contact Info

Product

Resources

About