Package elki.clustering.em

Class BetulaGMM

java.lang.Object

elki.clustering.em.BetulaGMM

All Implemented Interfaces:

Algorithm, ClusteringAlgorithm<Clustering<EMModel>>

Direct Known Subclasses:

BetulaGMMWeighted

@Reference(authors="Andreas Lang and Erich Schubert",
           title="BETULA: Fast Clustering of Large Data with Improved BIRCH CF-Trees",
           booktitle="Information Systems",
           url="https://doi.org/10.1016/j.is.2021.101918",
           bibkey="DBLP:journals/is/LangS22")
public class BetulaGMM
extends java.lang.Object
implements ClusteringAlgorithm<Clustering<EMModel>>

Clustering by expectation maximization (EM-Algorithm), also known as Gaussian Mixture Modeling (GMM), with optional MAP regularization. This version uses the BIRCH cluster feature centers only for responsibility estimation; the CF variances are only used for computing the models.

Reference:

Andreas Lang and Erich Schubert
BETULA: Fast Clustering of Large Data with Improved BIRCH CF-Trees
Information Systems

Since:

0.8.0

Author:

Andreas Lang

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	BetulaGMM.Par	Parameterizer

Nested classes/interfaces inherited from interface elki.Algorithm

Algorithm.Utils

Field Summary

Fields

Modifier and Type	Field	Description
(package private) CFTree.Factory<?>	cffactory	CFTree factory.
private double	delta	Delta parameter
(package private) BetulaClusterModelFactory<?>	initializer	Maximum number of iterations.
(package private) int	k	Number of cluster centers to initialize.
private static Logging	LOG	Class logger.
(package private) int	maxiter	Maximum number of iterations.
protected static double	MIN_LOGLIKELIHOOD	Minimum loglikelihood to avoid -infinity.
private double	prior	Prior to enable MAP estimation (use 0 for MLE)
private boolean	soft	Retain soft assignments.
static SimpleTypeInformation<double[]>	SOFT_TYPE	Soft assignment result type.

Constructor Summary

Constructors

Constructor	Description
BetulaGMM(CFTree.Factory<?> cffactory, double delta, int k, int maxiter, boolean soft, BetulaClusterModelFactory<?> initialization, double prior)	Constructor.

Method Summary

Modifier and Type	Method	Description
double	assignProbabilitiesToInstances(Relation<? extends NumberVector> relation, java.util.List<? extends BetulaClusterModel> models, WritableDataStore<double[]> probClusterIGivenX)	Assigns the current probability values to the instances in the database and compute the expectation value of the current mixture of distributions.
double	assignProbabilitiesToInstances(java.util.ArrayList<? extends ClusterFeature> cfs, java.util.List<? extends BetulaClusterModel> models, java.util.Map<ClusterFeature,double[]> probClusterIGivenX)	Assigns the current probability values to the instances in the database and compute the expectation value of the current mixture of distributions.
TypeInformation[]	getInputTypeRestriction()	Get the input type restriction used for negotiating the data query.
private boolean	isSoft()
void	recomputeCovarianceMatrices(java.util.ArrayList<? extends ClusterFeature> cfs, java.util.Map<ClusterFeature,double[]> probClusterIGivenX, java.util.List<? extends BetulaClusterModel> models, double prior, int n)	Recompute the covariance matrixes.
Clustering<EMModel>	run(Relation<NumberVector> relation)	Run the clustering algorithm.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface elki.clustering.ClusteringAlgorithm

autorun

Field Detail

LOG

private static final Logging LOG

Class logger.

cffactory

CFTree.Factory<?> cffactory

CFTree factory.

k

int k

Number of cluster centers to initialize.

delta

private double delta

Delta parameter

maxiter

int maxiter

Maximum number of iterations.

prior

private double prior

Prior to enable MAP estimation (use 0 for MLE)

soft

private boolean soft

Retain soft assignments.

MIN_LOGLIKELIHOOD

protected static final double MIN_LOGLIKELIHOOD

Minimum loglikelihood to avoid -infinity.

See Also:

Constant Field Values

SOFT_TYPE

public static final SimpleTypeInformation<double[]> SOFT_TYPE

Soft assignment result type.

initializer

BetulaClusterModelFactory<?> initializer

Maximum number of iterations.

Constructor Detail

BetulaGMM

public BetulaGMM(CFTree.Factory<?> cffactory,
                 double delta,
                 int k,
                 int maxiter,
                 boolean soft,
                 BetulaClusterModelFactory<?> initialization,
                 double prior)

Constructor.

Parameters:

cffactory - CFTree factory

k - Number of clusters

maxiter - Maximum number of iterations

soft - Return soft clustering results

initialization - Initialization method

prior - MAP prior

Method Detail

getInputTypeRestriction

public TypeInformation[] getInputTypeRestriction()

Description copied from interface: Algorithm

Get the input type restriction used for negotiating the data query.

Specified by:

getInputTypeRestriction in interface Algorithm

Returns:

Type restriction

run

public Clustering<EMModel> run(Relation<NumberVector> relation)

Run the clustering algorithm.

Parameters:

relation - Input data

Returns:

Clustering

isSoft

private boolean isSoft()

assignProbabilitiesToInstances

public double assignProbabilitiesToInstances(java.util.ArrayList<? extends ClusterFeature> cfs,
                                             java.util.List<? extends BetulaClusterModel> models,
                                             java.util.Map<ClusterFeature,double[]> probClusterIGivenX)

Assigns the current probability values to the instances in the database and compute the expectation value of the current mixture of distributions.

Computed as the sum of the logarithms of the prior probability of each instance.

Parameters:

cfs - the cluster features to evaluate

models - Cluster models

probClusterIGivenX - Output storage for cluster probabilities

Returns:

the expectation value of the current mixture of distributions

assignProbabilitiesToInstances

public double assignProbabilitiesToInstances(Relation<? extends NumberVector> relation,
                                             java.util.List<? extends BetulaClusterModel> models,
                                             WritableDataStore<double[]> probClusterIGivenX)

Assigns the current probability values to the instances in the database and compute the expectation value of the current mixture of distributions.

Computed as the sum of the logarithms of the prior probability of each instance.

Parameters:

relation - the database used for assignment to instances

models - Cluster models

probClusterIGivenX - Output storage for cluster probabilities

Returns:

the expectation value of the current mixture of distributions

recomputeCovarianceMatrices

public void recomputeCovarianceMatrices(java.util.ArrayList<? extends ClusterFeature> cfs,
                                        java.util.Map<ClusterFeature,double[]> probClusterIGivenX,
                                        java.util.List<? extends BetulaClusterModel> models,
                                        double prior,
                                        int n)

Recompute the covariance matrixes.

Parameters:

cfs - Cluster features to evaluate

probClusterIGivenX - Object probabilities

models - Cluster models to update

prior - MAP prior (use 0 for MLE)

n - data set size