de.lmu.ifi.dbs.elki.algorithm.clustering.kmeans.quality

Interface KMeansQualityMeasure<O extends NumberVector>

Type Parameters:

O - Input Object restriction type

All Known Implementing Classes:

AbstractKMeansQualityMeasure, AkaikeInformationCriterion, BayesianInformationCriterion, BayesianInformationCriterionZhao, WithinClusterMeanDistanceQualityMeasure, WithinClusterVarianceQualityMeasure

public interface KMeansQualityMeasure<O extends NumberVector>

Interface for computing the quality of a K-Means clustering. Important note: some measures are ascending, others are descending!

Author:

Erich Schubert

Method Summary

Methods

Modifier and Type	Method and Description
boolean	ascending() Use ascending or descending ordering.
boolean	isBetter(double currentCost, double bestCost) Compare two scores.
<V extends O> double	quality(Clustering<? extends MeanModel> clustering, NumberVectorDistanceFunction<? super V> distanceFunction, Relation<V> relation) Calculates and returns the quality measure.

Method Detail

quality

<V extends O> double quality(Clustering<? extends MeanModel> clustering,
                           NumberVectorDistanceFunction<? super V> distanceFunction,
                           Relation<V> relation)

Calculates and returns the quality measure.

Type Parameters:

V - Actual vector type (could be a subtype of O!)

Parameters:

clustering - Clustering to analyze

distanceFunction - Distance function to use (usually Euclidean or squared Euclidean!)

relation - Relation for accessing objects

Returns:

quality measure

ascending

boolean ascending()

Use ascending or descending ordering.

Returns:

true when larger scores are better.

isBetter

boolean isBetter(double currentCost,
               double bestCost)

Compare two scores.

Parameters:

currentCost - New (candiate) cost/score

bestCost - Existing best cost/score (may be NaN)

Returns:

true when the new score is better, or the old score is NaN.