Package org.carrot2.math.matrix

Class KMeansMatrixFactorizationFactory

java.lang.Object

org.carrot2.attrs.AttrComposite

org.carrot2.math.matrix.IterativeMatrixFactorizationFactory

org.carrot2.math.matrix.KMeansMatrixFactorizationFactory

All Implemented Interfaces:

AcceptingVisitor, MatrixFactorizationFactory

public class KMeansMatrixFactorizationFactory
extends IterativeMatrixFactorizationFactory

Performs matrix factorization using the k-means clustering algorithm. This kind of factorization is sometimes referred to as Concept Decomposition Factorization.

Field Summary

Fields inherited from class org.carrot2.math.matrix.IterativeMatrixFactorizationFactory

DEFAULT_K, DEFAULT_MAX_ITERATIONS, DEFAULT_ORDERED, DEFAULT_SEEDING_FACTORY, DEFAULT_STOP_THRESHOLD, factorizationQuality, k, maxIterations, ordered, seedingFactory, stopThreshold

Fields inherited from class org.carrot2.attrs.AttrComposite

attributes

Constructor Summary

Constructors

Constructor	Description
KMeansMatrixFactorizationFactory()

Method Summary

Modifier and Type	Method	Description
MatrixFactorization	factorize(org.carrot2.math.mahout.matrix.DoubleMatrix2D A)	Factorizes matrix A.

Methods inherited from class org.carrot2.math.matrix.IterativeMatrixFactorizationFactory

createSeedingStrategy, estimateIterationsNumber, getK, getMaxIterations, getSeedingFactory, getStopThreshold, isOrdered, setK, setMaxIterations, setOrdered, setSeedingFactory, setStopThreshold

Methods inherited from class org.carrot2.attrs.AttrComposite

accept

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.carrot2.attrs.AcceptingVisitor

accept

Constructor Details

KMeansMatrixFactorizationFactory

public KMeansMatrixFactorizationFactory()

Method Details

factorize

public MatrixFactorization factorize(org.carrot2.math.mahout.matrix.DoubleMatrix2D A)

Description copied from interface: MatrixFactorizationFactory

Factorizes matrix A.

Parameters:

A - matrix to be factorized.