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

Class KMedoidsFastPAM1<V>

java.lang.Object

de.lmu.ifi.dbs.elki.algorithm.AbstractAlgorithm<R>

de.lmu.ifi.dbs.elki.algorithm.AbstractDistanceBasedAlgorithm<V,Clustering<MedoidModel>>

de.lmu.ifi.dbs.elki.algorithm.clustering.kmeans.KMedoidsPAM<V>

de.lmu.ifi.dbs.elki.algorithm.clustering.kmeans.KMedoidsFastPAM1<V>

Type Parameters:

V - vector datatype

All Implemented Interfaces:

Algorithm, ClusteringAlgorithm<Clustering<MedoidModel>>, DistanceBasedAlgorithm<V>

Direct Known Subclasses:

KMedoidsFastPAM

@Reference(authors="Erich Schubert, Peter J. Rousseeuw",
           title="Faster k-Medoids Clustering: Improving the PAM, CLARA, and CLARANS Algorithms",
           booktitle="preprint, to appear",
           url="https://arxiv.org/abs/1810.05691",
           bibkey="DBLP:journals/corr/abs-1810-05691")
 @Priority(value=-100)
public class KMedoidsFastPAM1<V>
extends KMedoidsPAM<V>

FastPAM1: A version of PAM that is O(k) times faster, i.e., now in O((n-k)²). The change here feels pretty small - we handle all k medoids in parallel using an array. But this means the innermost loop only gets executed in O(1/k) of all iterations, and thus we benefit on average.

This acceleration gives exactly (assuming perfect numerical accuracy) the same results as the original PAM. For further improvements that can affect the result, see also KMedoidsFastPAM, which is recommended for usage in practice.

Reference:

Erich Schubert, Peter J. Rousseeuw
Faster k-Medoids Clustering: Improving the PAM, CLARA, and CLARANS Algorithms
preprint, to appear

Since:

0.7.5

Author:

Erich Schubert

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
protected static class	KMedoidsFastPAM1.Instance Instance for a single dataset.
static class	KMedoidsFastPAM1.Parameterizer<V> Parameterization class.

Field Summary

Fields

Modifier and Type	Field and Description
private static java.lang.String	KEY Key for statistics logging.
private static Logging	LOG The logger for this class.

Fields inherited from class de.lmu.ifi.dbs.elki.algorithm.clustering.kmeans.KMedoidsPAM

initializer, k, maxiter

Fields inherited from class de.lmu.ifi.dbs.elki.algorithm.AbstractAlgorithm

ALGORITHM_ID

Fields inherited from interface de.lmu.ifi.dbs.elki.algorithm.DistanceBasedAlgorithm

DISTANCE_FUNCTION_ID

Constructor Summary

Constructors

Constructor and Description
KMedoidsFastPAM1(DistanceFunction<? super V> distanceFunction, int k, int maxiter, KMedoidsInitialization<V> initializer) Constructor.

Method Summary

Modifier and Type	Method and Description
protected Logging	getLogger() Get the (STATIC) logger for this class.
protected void	run(DistanceQuery<V> distQ, DBIDs ids, ArrayModifiableDBIDs medoids, WritableIntegerDataStore assignment) Run the main algorithm.

Methods inherited from class de.lmu.ifi.dbs.elki.algorithm.clustering.kmeans.KMedoidsPAM

getInputTypeRestriction, initialMedoids, run

Methods inherited from class de.lmu.ifi.dbs.elki.algorithm.AbstractDistanceBasedAlgorithm

getDistanceFunction

Methods inherited from class de.lmu.ifi.dbs.elki.algorithm.AbstractAlgorithm

run

Methods inherited from class java.lang.Object

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

Methods inherited from interface de.lmu.ifi.dbs.elki.algorithm.clustering.ClusteringAlgorithm

run

Field Detail

LOG

private static final Logging LOG

The logger for this class.

KEY

private static final java.lang.String KEY

Key for statistics logging.

Constructor Detail

KMedoidsFastPAM1

public KMedoidsFastPAM1(DistanceFunction<? super V> distanceFunction,
                        int k,
                        int maxiter,
                        KMedoidsInitialization<V> initializer)

Constructor.

Parameters:

distanceFunction - distance function

k - k parameter

maxiter - Maxiter parameter

initializer - Function to generate the initial means

Method Detail

run

protected void run(DistanceQuery<V> distQ,
                   DBIDs ids,
                   ArrayModifiableDBIDs medoids,
                   WritableIntegerDataStore assignment)

Description copied from class: KMedoidsPAM

Run the main algorithm. Internal use, for easier subclassing, this primarily is a wrapper around "new Instance" for subclasses.

Overrides:

run in class KMedoidsPAM<V>

Parameters:

distQ - Distance query

ids - IDs to process

medoids - Current medoids

assignment - Cluster assignment output

getLogger

protected Logging getLogger()

Description copied from class: AbstractAlgorithm

Get the (STATIC) logger for this class.

Overrides:

getLogger in class KMedoidsPAM<V>

Returns:

the static logger