Package elki.evaluation.clustering.internal

Class DaviesBouldinIndex

java.lang.Object

elki.evaluation.clustering.internal.DaviesBouldinIndex

All Implemented Interfaces:

Evaluator, ResultProcessor

@Reference(authors="D. L. Davies, D. W. Bouldin",
           title="A Cluster Separation Measure",
           booktitle="IEEE Transactions Pattern Analysis and Machine Intelligence 1(2)",
           url="https://doi.org/10.1109/TPAMI.1979.4766909",
           bibkey="DBLP:journals/pami/DaviesB79")
public class DaviesBouldinIndex
extends java.lang.Object
implements Evaluator

Compute the Davies-Bouldin index of a data set.

Reference:

D. L. Davies, D. W. Bouldin
A Cluster Separation Measure
IEEE Transactions Pattern Analysis and Machine Intelligence 1(2)

Since:

0.7.0

Author:

Stephan Baier

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	DaviesBouldinIndex.Par	Parameterization class.

Field Summary

Fields

Modifier and Type	Field	Description
private NumberVectorDistance<?>	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.
private double	p	Exponent p for computing the power mean.

Constructor Summary

Constructors

Constructor	Description
DaviesBouldinIndex(NumberVectorDistance<?> distance, NoiseHandling noiseOpt, double p)	Constructor.

Method Summary

Modifier and Type	Method	Description
double	evaluateClustering(Relation<? extends NumberVector> rel, Clustering<?> c)	Evaluate a single clustering.
void	processNewResult(java.lang.Object result)	Process a result.
double[]	withinGroupDistances(Relation<? extends NumberVector> rel, java.util.List<? extends Cluster<?>> clusters, NumberVector[] centroids)

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 NumberVectorDistance<?> distance

Distance function to use.

p

private double p

Exponent p for computing the power mean.

key

private java.lang.String key

Key for logging statistics.

Constructor Detail

DaviesBouldinIndex

public DaviesBouldinIndex(NumberVectorDistance<?> distance,
                          NoiseHandling noiseOpt,
                          double p)

Constructor.

Parameters:

distance - Distance function

noiseOpt - Flag to control noise handling

p - Power mean exponent p

Method Detail

evaluateClustering

public double evaluateClustering(Relation<? extends NumberVector> rel,
                                 Clustering<?> c)

Evaluate a single clustering.

Parameters:

rel - Data relation

c - Clustering

Returns:

DB-index

withinGroupDistances

public double[] withinGroupDistances(Relation<? extends NumberVector> rel,
                                     java.util.List<? extends Cluster<?>> clusters,
                                     NumberVector[] centroids)

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.