Gaussian Mixture Models

Cluster based on Gaussian mixture models using the Expectation-Maximization algorithm

Gaussian mixture models (GMMs) assign each observation to a cluster by maximizing the posterior probability that a data point belongs to its assigned cluster. Create a GMM object gmdistribution by fitting a model to data (fitgmdist) or by specifying parameter values (gmdistribution). Then, use object functions to perform cluster analysis (cluster, posterior, mahal), evaluate the model (cdf, pdf), and generate random variates (random).

Functions

expand all

fitgmdistFit Gaussian mixture model to data
gmdistributionCreate Gaussian mixture model
cdfCumulative distribution function for Gaussian mixture distribution
clusterConstruct clusters from Gaussian mixture distribution
mahalMahalanobis distance to Gaussian mixture component
pdfProbability density function for Gaussian mixture distribution
posteriorPosterior probability of Gaussian mixture component
randomRandom variate from Gaussian mixture distribution

Topics

Cluster Using Gaussian Mixture Model

Partition data into clusters with different sizes and correlation structures.

Cluster Gaussian Mixture Data Using Hard Clustering

Implement hard clustering on simulated data from a mixture of Gaussian distributions.

Cluster Gaussian Mixture Data Using Soft Clustering

Implement soft clustering on simulated data from a mixture of Gaussian distributions.

Tune Gaussian Mixture Models

Determine the best Gaussian mixture model (GMM) fit by adjusting the number of components and the component covariance matrix structure.

Statistics and Machine Learning Toolbox Documentation
Support
Mastering Machine Learning: A Step-by-Step Guide with MATLAB

Mastering Machine Learning: A Step-by-Step Guide with MATLAB

Download ebook