Package elki.clustering.silhouette

Class FasterMSC<O>

java.lang.Object

elki.clustering.kmedoids.PAM<O>

elki.clustering.silhouette.PAMSIL<O>

elki.clustering.silhouette.PAMMEDSIL<O>

elki.clustering.silhouette.FastMSC<O>

elki.clustering.silhouette.FasterMSC<O>

Type Parameters:

O -

All Implemented Interfaces:

Algorithm, ClusteringAlgorithm<Clustering<MedoidModel>>, KMedoidsClustering<O>

@Reference(authors="Lars Lenssen and Erich Schubert",
           title="Clustering by Direct Optimization of the Medoid Silhouette",
           booktitle="Int. Conf. on Similarity Search and Applications, SISAP 2022",
           url="https://doi.org/10.1007/978-3-031-17849-8_15",
           bibkey="DBLP:conf/sisap/LenssenS22")
@Priority(200)
public class FasterMSC<O>
extends FastMSC<O>

Fast and Eager Medoid Silhouette Clustering.

This clustering algorithm tries to find an optimal silhouette clustering for an approximation to the silhouette called "medoid silhouette" using a swap-based heuristic similar to PAM. By also caching the distance to the third nearest center (compare to FastPAM, which only used the second nearest), we are able to reduce the runtime per iteration to just O(n²), which yields an acceptable run time for many use cases, while often finding a solution with better silhouette than other clustering methods. This version also performs eager swapping instead of a steepest descent, i.e., it performs any swap that improves the medoid silhouette immediately, and hence may need fewer iterations.

Reference:

Lars Lenssen and Erich Schubert
Clustering by Direct Optimization of the Medoid Silhouette
Int. Conf. on Similarity Search and Applications, SISAP 2022

Author:

Erich Schubert

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
protected class	FasterMSC.Instance	FasterMSC clustering instance for a particular data set.
protected class	FasterMSC.Instance2	FasterMSC clustering instance for k=2, simplified.
static class	FasterMSC.Par<O>	Parameterization class.

Nested classes/interfaces inherited from class elki.clustering.silhouette.FastMSC

FastMSC.Record

Nested classes/interfaces inherited from interface elki.Algorithm

Algorithm.Utils

Field Summary

Fields

Modifier and Type	Field	Description
private static Logging	LOG	The logger for this class.

Fields inherited from class elki.clustering.kmedoids.PAM

distance, initializer, k, maxiter

Constructor Summary

Constructors

Constructor	Description
FasterMSC(Distance<? super O> distance, int k, int maxiter, KMedoidsInitialization<O> initializer)	Constructor.

Method Summary

Modifier and Type	Method	Description
protected Logging	getLogger()	Get the static class logger.
Clustering<MedoidModel>	run(Relation<O> relation, int k, DistanceQuery<? super O> distQ)	Run k-medoids clustering with a given distance query. Not a very elegant API, but needed for some types of nested k-medoids.

Methods inherited from class elki.clustering.silhouette.FastMSC

loss

Methods inherited from class elki.clustering.kmedoids.PAM

getInputTypeRestriction, initialMedoids, run, wrapResult

Methods inherited from class java.lang.Object

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

Methods inherited from interface elki.clustering.ClusteringAlgorithm

autorun

Field Detail

LOG

private static final Logging LOG

The logger for this class.

Constructor Detail

FasterMSC

public FasterMSC(Distance<? super O> distance,
                 int k,
                 int maxiter,
                 KMedoidsInitialization<O> initializer)

Constructor.

Parameters:

distance - Distance function

k - Number of cluster

maxiter - Maximum number of iterations

initializer - Initialization

Method Detail

run

public Clustering<MedoidModel> run(Relation<O> relation,
                                   int k,
                                   DistanceQuery<? super O> distQ)

Description copied from interface: KMedoidsClustering

Run k-medoids clustering with a given distance query.
Not a very elegant API, but needed for some types of nested k-medoids.

Specified by:

run in interface KMedoidsClustering<O>

Overrides:

run in class FastMSC<O>

Parameters:

relation - relation to use

k - Number of clusters

distQ - Distance query to use

Returns:

result

getLogger

protected Logging getLogger()

Description copied from class: PAM

Get the static class logger.

Overrides:

getLogger in class FastMSC<O>