jminhep.algorithms

Class FuzzyCMeansAlg

java.lang.Object

jminhep.algorithms.FuzzyCMeansAlg

public class FuzzyCMeansAlg
extends Object

Fuzzy C-means cluster algorithm. Look at details in: http://home.dei.polimi.it/matteucc/Clustering/tutorial_html/cmeans.html

Constructor Summary

Constructors

Constructor and Description
FuzzyCMeansAlg(DataHolder data) Initialize Fuzzy C-means calculations

Method Summary

Modifier and Type	Method and Description
void	delete() Clear
int[]	getAssignments() Returns cluster assignments of all points.
DataHolder	getCenters() Returns cluster centers
int	getClusters() Get the number of clusters
double	getCompactness() This method returns the Compactness and Separation measure of cluster validity (see Fuzzy Algorithms With Applications to Image Processing and Pattern Recognition, Zheru Chi, Hong Yan, Tuan Pham, World Scientific, pp.
int	getError() Get error code.
double[][]	getMembeship() Return cluster membership
String	getName() Get the description of the algorithm
int[]	getNumberPoints() Returns the number of points in each cluster.
double	getPartitionCoefficient() This method returns the Partition Coefficient measure of cluster validity (see Fuzzy Algorithms With Applications to Image Processing and Pattern Recognition, Zheru Chi, Hong Yan, Tuan Pham, World Scientific, pp.
double	getPartitionEntropy() This method returns the Partition Entropy measure of cluster validity (see Fuzzy Algorithms With Applications to Image Processing and Pattern Recognition, Zheru Chi, Hong Yan, Tuan Pham, World Scientific, pp.
long	getPosition() This method returns a measure of the progress of the algorithm.
long	getSize() This method returns the estimated size (steps) for this task.
boolean	isFinished() This method returns true if the clustering has finished.
void	run() Run classic Fuzzy C-Means clustering algorithm: Calculate the cluster centers.
void	runBest() Runs for the best estimate
void	setClusters(int N) Set number of clusters for calculations.
void	setOptions(int maxIterations, double epsilon, double fuzziness) Set initial conditions for clustering.
void	setProb(double probClusters) Set probability of associations with each cluster.

Methods inherited from class java.lang.Object

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

Constructor Detail

FuzzyCMeansAlg

public FuzzyCMeansAlg(DataHolder data)

Initialize Fuzzy C-means calculations

Parameters:

d - input data

Method Detail

setClusters

public void setClusters(int N)

Set number of clusters for calculations.

Parameters:

numClusters - Set the desired number of clusters.

getClusters

public int getClusters()

Get the number of clusters

Returns:

number of clusters

getMembeship

public double[][] getMembeship()

Return cluster membership

Returns:

membership matrix (2D)

setOptions

public void setOptions(int maxIterations,
                       double epsilon,
                       double fuzziness)

Set initial conditions for clustering.

Parameters:

maxIterations - the maximum number of iterations.

fuzziness - the fuzziness (a.k.a. the "m" value)

epsilon - a small value used to verify if clustering has converged.

delete

public void delete()

Clear

setProb

public void setProb(double probClusters)

Set probability of associations with each cluster. The default value is 0.68. Does dot affect calculations.

Parameters:

probClusters - probability association.

getCenters

public DataHolder getCenters()

Returns cluster centers

Returns:

DataHolder with cluster centers

getAssignments

public int[] getAssignments()

Returns cluster assignments of all points. Affected by the method setProb(), which sets association probability (default is 0.68).

Returns:

array with cluster assignments

getNumberPoints

public int[] getNumberPoints()

Returns the number of points in each cluster. Affected by the method setProb(), which sets association probability (default is 0.68). Look also at getAssignment() and getMembership()

Returns:

number of points in each cluster

runBest

public void runBest()

Runs for the best estimate

run

public void run()

Run classic Fuzzy C-Means clustering algorithm: Calculate the cluster centers. Update the membership function. Calculate statistics and repeat from 1 if needed.

getSize

public long getSize()

This method returns the estimated size (steps) for this task. The value is, of course, an approximation, just so we will be able to give the user a feedback on the processing time. In this case, the value is calculated as the number of loops in the run() method.

getPosition

public long getPosition()

This method returns a measure of the progress of the algorithm.

isFinished

public boolean isFinished()

This method returns true if the clustering has finished.

getPartitionCoefficient

public double getPartitionCoefficient()

This method returns the Partition Coefficient measure of cluster validity (see Fuzzy Algorithms With Applications to Image Processing and Pattern Recognition, Zheru Chi, Hong Yan, Tuan Pham, World Scientific, pp. 91)

getPartitionEntropy

public double getPartitionEntropy()

This method returns the Partition Entropy measure of cluster validity (see Fuzzy Algorithms With Applications to Image Processing and Pattern Recognition, Zheru Chi, Hong Yan, Tuan Pham, World Scientific, pp. 91)

getError

public int getError()

Get error code.

Returns:

if 0, clustering looks OK

getCompactness

public double getCompactness()

This method returns the Compactness and Separation measure of cluster validity (see Fuzzy Algorithms With Applications to Image Processing and Pattern Recognition, Zheru Chi, Hong Yan, Tuan Pham, World Scientific, pp. 93)

getName

public String getName()

Get the description of the algorithm

Returns:

algorithm description