Package elki.clustering.kmeans.initialization

Interface KMeansInitialization

All Known Implementing Classes:

AbstractKMeansInitialization, AFKMC2, BUILD, FarthestPoints, FarthestSumPoints, FirstK, KMC2, KMeansPlusPlus, LAB, Ostrovsky, ParkJun, Predefined, RandomlyChosen, RandomNormalGenerated, RandomUniformGenerated, SampleKMeans, SphericalAFKMC2, SphericalKMeansPlusPlus

public interface KMeansInitialization

Interface for initializing K-Means

Since:

0.5.0

Author:

Erich Schubert

Method Summary

Modifier and Type	Method	Description
double[][]	chooseInitialMeans(Relation<? extends NumberVector> relation, int k, NumberVectorDistance<?> distance)	Choose initial means

Method Detail

chooseInitialMeans

double[][] chooseInitialMeans(Relation<? extends NumberVector> relation,
                              int k,
                              NumberVectorDistance<?> distance)

Choose initial means

Parameters:

relation - Relation

k - Parameter k

distance - Distance function

Returns:

List of chosen means for k-means