Package elki.evaluation.clustering.internal

Class CIndex<O>

java.lang.Object

elki.evaluation.clustering.internal.CIndex<O>

All Implemented Interfaces:

Evaluator, ResultProcessor

@Reference(authors="L. J. Hubert, J. R. Levin",
           title="A general statistical framework for assessing categorical clustering in free recall",
           booktitle="Psychological Bulletin, Vol. 83(6)",
           url="https://doi.org/10.1037/0033-2909.83.6.1072",
           bibkey="doi:10.1037/0033-2909.83.6.1072")
public class CIndex<O>
extends java.lang.Object
implements Evaluator

Compute the C-index of a data set.

Note: This requires pairwise distance computations, so it is not recommended to use this on larger data sets.

Reference:

L. J. Hubert, J. R. Levin
A general statistical framework for assessing categorical clustering in free recall
Psychological Bulletin, Vol. 83(6)

Since:

0.7.0

Author:

Stephan Baier, Erich Schubert

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	CIndex.Par<O>	Parameterization class.

Field Summary

Fields

Modifier and Type	Field	Description
private Distance<? super O>	distance	Distance function to use.
private java.lang.String	key	Key for logging statistics.
private static Logging	LOG	Logger for debug output.
private NoiseHandling	noiseOption	Option for noise handling.

Constructor Summary

Constructors

Constructor	Description
CIndex(Distance<? super O> distance, NoiseHandling noiseOpt)	Constructor.

Method Summary

Modifier and Type	Method	Description
double	evaluateClustering(Relation<? extends O> rel, DistanceQuery<O> dq, Clustering<?> c)	Evaluate a single clustering.
protected double	processCluster(Cluster<?> cluster, java.util.List<? extends Cluster<?>> clusters, int i, DistanceQuery<O> dq, DoubleHeap maxDists, DoubleHeap minDists, int w)
void	processNewResult(java.lang.Object result)	Process a result.
protected void	processSingleton(Cluster<?> cluster, Relation<? extends O> rel, DistanceQuery<O> dq, DoubleHeap maxDists, DoubleHeap minDists, int w)

Methods inherited from class java.lang.Object

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

Field Detail

LOG

private static final Logging LOG

Logger for debug output.

noiseOption

private NoiseHandling noiseOption

Option for noise handling.

distance

private Distance<? super O> distance

Distance function to use.

key

private java.lang.String key

Key for logging statistics.

Constructor Detail

CIndex

public CIndex(Distance<? super O> distance,
              NoiseHandling noiseOpt)

Constructor.

Parameters:

distance - Distance function

noiseOpt - Flag to control noise handling

Method Detail

evaluateClustering

public double evaluateClustering(Relation<? extends O> rel,
                                 DistanceQuery<O> dq,
                                 Clustering<?> c)

Evaluate a single clustering.

Parameters:

rel - Data relation

c - Clustering

Returns:

C-Index

processCluster

protected double processCluster(Cluster<?> cluster,
                                java.util.List<? extends Cluster<?>> clusters,
                                int i,
                                DistanceQuery<O> dq,
                                DoubleHeap maxDists,
                                DoubleHeap minDists,
                                int w)

processSingleton

protected void processSingleton(Cluster<?> cluster,
                                Relation<? extends O> rel,
                                DistanceQuery<O> dq,
                                DoubleHeap maxDists,
                                DoubleHeap minDists,
                                int w)

processNewResult

public void processNewResult(java.lang.Object result)

Description copied from interface: ResultProcessor

Process a result.

Specified by:

processNewResult in interface ResultProcessor

Parameters:

result - Newly added result subtree.