Package elki.clustering.kmeans.quality

Interface KMeansQualityMeasure<O extends elki.data.NumberVector>

Type Parameters:

O - Input Object restriction type

All Known Implementing Classes:

AbstractKMeansQualityMeasure, AkaikeInformationCriterion, AkaikeInformationCriterionXMeans, BayesianInformationCriterion, BayesianInformationCriterionXMeans, BayesianInformationCriterionZhao, WithinClusterMeanDistance, WithinClusterVariance

public interface KMeansQualityMeasure<O extends elki.data.NumberVector>

Interface for computing the quality of a K-Means clustering.

Important note: some measures are ascending, others are descending, so use the method isBetter(double, double) for ordering.

Since:

0.6.0

Author:

Erich Schubert

Method Summary

Modifier and Type	Method	Description
boolean	isBetter(double currentCost, double bestCost)	Compare two scores.
<V extends O> double	quality(Clustering<? extends MeanModel> clustering, elki.distance.NumberVectorDistance<? super V> distance, elki.database.relation.Relation<V> relation)	Calculates and returns the quality measure.

Method Detail

quality

<V extends O> double quality(Clustering<? extends MeanModel> clustering,
                             elki.distance.NumberVectorDistance<? super V> distance,
                             elki.database.relation.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

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

relation - Relation for accessing objects

Returns:

quality measure

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.