Package elki.clustering.hierarchical

Class AGNES.Instance

java.lang.Object

elki.clustering.hierarchical.AGNES.Instance

Direct Known Subclasses:

Anderberg.Instance, MedoidLinkage.Instance, MiniMax.Instance, NNChain.Instance

Enclosing class:

AGNES<O>

public static class AGNES.Instance
extends java.lang.Object

Main worker instance of AGNES.

Author:

Erich Schubert

Field Summary

Fields

Modifier and Type	Field	Description
protected ClusterMergeHistoryBuilder	builder	Cluster result builder
protected int	end	Active set size
protected Linkage	linkage	Current linkage method in use.
protected ClusterDistanceMatrix	mat	Cluster distance matrix

Constructor Summary

Constructors

Constructor	Description
Instance(Linkage linkage)	Constructor.

Method Summary

Modifier and Type	Method	Description
protected int	findMerge()	Perform the next merge step in AGNES.
protected void	merge(double mindist, int x, int y)	Execute the cluster merge.
ClusterMergeHistory	run(ClusterDistanceMatrix mat, ClusterMergeHistoryBuilder builder)	Run the main algorithm.
protected static int	shrinkActiveSet(int[] clustermap, int end, int x)	Shrink the active set: if the last x objects are all merged, we can reduce the working size accordingly.
protected void	updateMatrix(double mindist, int x, int y, int sizex, int sizey)	Update the scratch distance matrix.

Methods inherited from class java.lang.Object

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

Field Detail

linkage

protected Linkage linkage

Current linkage method in use.

mat

protected ClusterDistanceMatrix mat

Cluster distance matrix

builder

protected ClusterMergeHistoryBuilder builder

Cluster result builder

end

protected int end

Active set size

Constructor Detail

Instance

public Instance(Linkage linkage)

Constructor.

Parameters:

linkage - Linkage

Method Detail

run

public ClusterMergeHistory run(ClusterDistanceMatrix mat,
                               ClusterMergeHistoryBuilder builder)

Run the main algorithm.

Parameters:

mat - Distance matrix

builder - Result builder

Returns:

Cluster history

findMerge

protected int findMerge()

Perform the next merge step in AGNES.

Returns:

the index that has disappeared, for shrinking the working set

merge

protected void merge(double mindist,
                     int x,
                     int y)

Execute the cluster merge.

Parameters:

mindist - Distance that was used for merging

x - First matrix position

y - Second matrix position

updateMatrix

protected void updateMatrix(double mindist,
                            int x,
                            int y,
                            int sizex,
                            int sizey)

Update the scratch distance matrix.

Parameters:

mindist - Minimum distance

x - First matrix position

y - Second matrix position

sizex - Old size of first cluster

sizey - Old size of second cluster

shrinkActiveSet

protected static int shrinkActiveSet(int[] clustermap,
                                     int end,
                                     int x)

Shrink the active set: if the last x objects are all merged, we can reduce the working size accordingly.

Parameters:

clustermap - Map to current clusters

end - Current active set size

x - Last merged object

Returns:

New active set size