net.sf.javaml.clustering
Class IterativeMultiKMeans

java.lang.Object
  net.sf.javaml.clustering.IterativeMultiKMeans

All Implemented Interfaces:

Clusterer

public class IterativeMultiKMeans
extends java.lang.Object
implements Clusterer

This class implements an extension of KMeans, combining Iterative- en MultiKMeans. SKM will be run several iterations with a different k value, starting from kMin and increasing to kMax, and several iterations for each k. Each clustering result is evaluated with an evaluation score, the result with the best score will be returned as final result. XXX add reference XXX add pseudo code

Author:

Thomas Abeel (thomas@abeel.be), Andreas De Rijcke

Constructor Summary

IterativeMultiKMeans(ClusterEvaluation ce)
default constructor

IterativeMultiKMeans(int kMin, int kMax, ClusterEvaluation ce)
XXX add doc

IterativeMultiKMeans(int kMin, int kMax, int iterations, int repeats, DistanceMeasure dm, ClusterEvaluation ce)
XXX add doc

Method Summary

Dataset[]

cluster(Dataset data) XXX add doc

Methods inherited from class java.lang.Object

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

Constructor Detail

IterativeMultiKMeans

public IterativeMultiKMeans(ClusterEvaluation ce)

default constructor

Parameters:

ClusterEvaluation - ce

IterativeMultiKMeans

public IterativeMultiKMeans(int kMin,
                            int kMax,
                            ClusterEvaluation ce)

XXX add doc

Parameters:

kMin -

kMax -

ClusterEvaluation - ce

IterativeMultiKMeans

public IterativeMultiKMeans(int kMin,
                            int kMax,
                            int iterations,
                            int repeats,
                            DistanceMeasure dm,
                            ClusterEvaluation ce)

XXX add doc

Parameters:

kMin -

kMax -

iterations -

repeats -

DistanceMeasure - dm

ClusterEvaluation - ce

Method Detail

cluster

public Dataset[] cluster(Dataset data)

XXX add doc

Specified by:

cluster in interface Clusterer

Parameters:

data - the data set on which to execute the clustering.

Returns:

the different clusters obtained by this clustering algorithm. Each cluster is represented as a separate data set.