Package de.lmu.ifi.dbs.elki.algorithm.clustering.kmeans

K-means clustering and variations.

See: Description

Interface Summary

Interface	Description
KMeans<V extends NumberVector<?>,D extends Distance<?>,M extends MeanModel<V>>	Some constants and options shared among kmeans family algorithms.
KMeansInitialization<V>	Interface for initializing K-Means
KMedoidsInitialization<V>	Interface for initializing K-Medoids.

Class Summary

Class	Description
AbstractKMeans<V extends NumberVector<?>,D extends Distance<D>,M extends MeanModel<V>>	Abstract base class for k-means implementations.
AbstractKMeans.Parameterizer<V extends NumberVector<?>,D extends Distance<D>>	Parameterization class.
AbstractKMeansInitialization<V>	Abstract base class for common k-means initializations.
AbstractKMeansInitialization.Parameterizer<V>	Parameterization class.
BestOfMultipleKMeans<V extends NumberVector<?>,D extends Distance<?>,M extends MeanModel<V>>	Run K-Means multiple times, and keep the best run.
BestOfMultipleKMeans.Parameterizer<V extends NumberVector<?>,D extends Distance<D>,M extends MeanModel<V>>	Parameterization class.
FarthestPointsInitialMeans<V,D extends NumberDistance<D,?>>	K-Means initialization by repeatedly choosing the farthest point.
FarthestPointsInitialMeans.Parameterizer<V,D extends NumberDistance<D,?>>	Parameterization class.
FirstKInitialMeans<V>	Initialize K-means by using the first k objects as initial means.
FirstKInitialMeans.Parameterizer<V extends NumberVector<?>>	Parameterization class.
KMeansBatchedLloyd<V extends NumberVector<?>,D extends Distance<D>>	Provides the k-means algorithm, using Lloyd-style bulk iterations.
KMeansBatchedLloyd.Parameterizer<V extends NumberVector<?>,D extends Distance<D>>	Parameterization class.
KMeansBisecting<V extends NumberVector<?>,D extends Distance<?>,M extends MeanModel<V>>	The bisecting k-means algorithm works by starting with an initial partitioning into two clusters, then repeated splitting of the largest cluster to get additional clusters.
KMeansBisecting.Parameterizer<V extends NumberVector<?>,D extends Distance<?>,M extends MeanModel<V>>	Parameterization class.
KMeansHybridLloydMacQueen<V extends NumberVector<?>,D extends Distance<D>>	Provides the k-means algorithm, alternating between MacQueen-style incremental processing and Lloyd-Style batch steps.
KMeansHybridLloydMacQueen.Parameterizer<V extends NumberVector<?>,D extends Distance<D>>	Parameterization class.
KMeansLloyd<V extends NumberVector<?>,D extends Distance<D>>	Provides the k-means algorithm, using Lloyd-style bulk iterations.
KMeansLloyd.Parameterizer<V extends NumberVector<?>,D extends Distance<D>>	Parameterization class.
KMeansMacQueen<V extends NumberVector<?>,D extends Distance<D>>	Provides the k-means algorithm, using MacQueen style incremental updates.
KMeansMacQueen.Parameterizer<V extends NumberVector<?>,D extends Distance<D>>	Parameterization class.
KMeansPlusPlusInitialMeans<V,D extends NumberDistance<D,?>>	K-Means++ initialization for k-means.
KMeansPlusPlusInitialMeans.Parameterizer<V,D extends NumberDistance<D,?>>	Parameterization class.
KMediansLloyd<V extends NumberVector<?>,D extends Distance<D>>	Provides the k-medians clustering algorithm, using Lloyd-style bulk iterations.
KMediansLloyd.Parameterizer<V extends NumberVector<?>,D extends Distance<D>>	Parameterization class.
KMedoidsEM<V,D extends NumberDistance<D,?>>	Provides the k-medoids clustering algorithm, using a "bulk" variation of the "Partitioning Around Medoids" approach.
KMedoidsEM.Parameterizer<V,D extends NumberDistance<D,?>>	Parameterization class.
KMedoidsPAM<V,D extends NumberDistance<D,?>>	Provides the k-medoids clustering algorithm, using the "Partitioning Around Medoids" approach.
KMedoidsPAM.Parameterizer<V,D extends NumberDistance<D,?>>	Parameterization class.
PAMInitialMeans<V,D extends NumberDistance<D,?>>	PAM initialization for k-means (and of course, PAM).
PAMInitialMeans.Parameterizer<V,D extends NumberDistance<D,?>>	Parameterization class.
RandomlyChosenInitialMeans<V>	Initialize K-means by randomly choosing k exsiting elements as cluster centers.
RandomlyChosenInitialMeans.Parameterizer<V>	Parameterization class.
RandomlyGeneratedInitialMeans<V extends NumberVector<?>>	Initialize k-means by generating random vectors (within the data sets value range).
RandomlyGeneratedInitialMeans.Parameterizer<V extends NumberVector<?>>	Parameterization class.
SampleKMeansInitialization<V extends NumberVector<?>,D extends Distance<?>>	Initialize k-means by running k-means on a sample of the data set only.
SampleKMeansInitialization.Parameterizer<V extends NumberVector<?>,D extends Distance<?>>	Parameterization class.

Package de.lmu.ifi.dbs.elki.algorithm.clustering.kmeans Description

K-means clustering and variations.