Package elki.clustering.hierarchical

Class ClusterPrototypeMergeHistory

java.lang.Object

elki.clustering.hierarchical.ClusterMergeHistory

elki.clustering.hierarchical.ClusterPrototypeMergeHistory

public class ClusterPrototypeMergeHistory
extends ClusterMergeHistory

Cluster merge history with additional cluster prototypes (for HACAM, MedoidLinkage, and MiniMax clustering)

Since:

0.8.0

Author:

Erich Schubert

Field Summary

Fields

Modifier and Type	Field	Description
protected ArrayDBIDs	prototypes	Cluster prototypes

Fields inherited from class elki.clustering.hierarchical.ClusterMergeHistory

distances, ids, isSquared, merges, positions, sizes

Constructor Summary

Constructors

Constructor	Description
ClusterPrototypeMergeHistory(ArrayDBIDs ids, int[] merges, double[] distances, int[] sizes, boolean isSquared, ArrayDBIDs prototypes)	Constructor.

Method Summary

Modifier and Type	Method	Description
DBID	prototype(int i)	Get the prototype of cluster i.

Methods inherited from class elki.clustering.hierarchical.ClusterMergeHistory

assignVar, getDBIDs, getMergeA, getMergeB, getMergeHeight, getPositions, getSize, isSquared, numMerges, size

Methods inherited from class java.lang.Object

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

Field Detail

prototypes

protected ArrayDBIDs prototypes

Cluster prototypes

Constructor Detail

ClusterPrototypeMergeHistory

public ClusterPrototypeMergeHistory(ArrayDBIDs ids,
                                    int[] merges,
                                    double[] distances,
                                    int[] sizes,
                                    boolean isSquared,
                                    ArrayDBIDs prototypes)

Constructor.

Parameters:

ids - Initial object ids

merges - Merge history 2*(N-1) values

distances - Distances

sizes - Cluster sizes

isSquared - If distances are squared distances

prototypes - Cluster prototypes

Method Detail

prototype

public DBID prototype(int i)

Get the prototype of cluster i.

Parameters:

i - Merge number

Returns:

Prototype