de.lmu.ifi.dbs.elki.algorithm.clustering.hierarchical.birch

Class AverageIntraclusterDistance

java.lang.Object

de.lmu.ifi.dbs.elki.algorithm.clustering.hierarchical.birch.AverageIntraclusterDistance

All Implemented Interfaces:

BIRCHDistance

@Alias(value="D3")
 @Reference(authors="T. Zhang",
           title="Data Clustering for Very Large Datasets Plus Applications",
           booktitle="University of Wisconsin Madison, Technical Report #1355",
           url="ftp://ftp.cs.wisc.edu/pub/techreports/1997/TR1355.pdf",
           bibkey="tr/wisc/Zhang97")
public class AverageIntraclusterDistance
extends java.lang.Object
implements BIRCHDistance

Average intracluster distance.

Reference:

Data Clustering for Very Large Datasets Plus Applications
T. Zhang
Doctoral Dissertation, 1997.

Note: this distance did not work well in the original work, apparently.

Since:

0.7.5

Author:

Erich Schubert

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	AverageIntraclusterDistance.Parameterizer Parameterization class.

Field Summary

Fields

Modifier and Type	Field and Description
static AverageIntraclusterDistance	STATIC Static instance.

Constructor Summary

Constructors

Constructor and Description
AverageIntraclusterDistance()

Method Summary

Modifier and Type	Method and Description
double	squaredDistance(ClusteringFeature cf1, ClusteringFeature cf2) Distance between two clustering features.
double	squaredDistance(NumberVector v, ClusteringFeature cf) Distance of a vector to a clustering feature.

Methods inherited from class java.lang.Object

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

Field Detail

STATIC

public static final AverageIntraclusterDistance STATIC

Static instance.

Constructor Detail

AverageIntraclusterDistance

public AverageIntraclusterDistance()

Method Detail

squaredDistance

public double squaredDistance(NumberVector v,
                              ClusteringFeature cf)

Description copied from interface: BIRCHDistance

Distance of a vector to a clustering feature.

Specified by:

squaredDistance in interface BIRCHDistance

Parameters:

v - Vector

cf - Clustering Feature

Returns:

Distance

squaredDistance

public double squaredDistance(ClusteringFeature cf1,
                              ClusteringFeature cf2)

Description copied from interface: BIRCHDistance

Distance between two clustering features.

Specified by:

squaredDistance in interface BIRCHDistance

Parameters:

cf1 - First clustering feature

cf2 - Second clustering feature

Returns:

Distance