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

Class EuclideanDistanceCriterion

java.lang.Object

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

All Implemented Interfaces:

BIRCHAbsorptionCriterion

public class EuclideanDistanceCriterion
extends java.lang.Object
implements BIRCHAbsorptionCriterion

Distance criterion. This is not found in the original work, but used in many implementation attempts: assign points if the Euclidean distances is below a threshold.

Since:

0.7.5

Author:

Erich Schubert

Nested Class Summary

Nested Classes

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

Field Summary

Fields

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

Constructor Summary

Constructors

Constructor and Description
EuclideanDistanceCriterion()

Method Summary

Modifier and Type	Method and Description
double	squaredCriterion(ClusteringFeature f1, ClusteringFeature f2) Quality when merging two CFs.
double	squaredCriterion(ClusteringFeature f1, NumberVector n) Quality of a CF when adding a data point

Methods inherited from class java.lang.Object

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

Field Detail

STATIC

public static final EuclideanDistanceCriterion STATIC

Static instance.

Constructor Detail

EuclideanDistanceCriterion

public EuclideanDistanceCriterion()

Method Detail

squaredCriterion

public double squaredCriterion(ClusteringFeature f1,
                               NumberVector n)

Description copied from interface: BIRCHAbsorptionCriterion

Quality of a CF when adding a data point

Specified by:

squaredCriterion in interface BIRCHAbsorptionCriterion

Parameters:

f1 - Clustering feature

n - Data point

Returns:

Quality

squaredCriterion

public double squaredCriterion(ClusteringFeature f1,
                               ClusteringFeature f2)

Description copied from interface: BIRCHAbsorptionCriterion

Quality when merging two CFs.

Specified by:

squaredCriterion in interface BIRCHAbsorptionCriterion

Parameters:

f1 - First clustering feature

f2 - Second clustering feature

Returns:

Quality