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

Class KMedoidsFastPAM1.Instance

java.lang.Object

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

de.lmu.ifi.dbs.elki.algorithm.clustering.kmeans.KMedoidsFastPAM1.Instance

Direct Known Subclasses:

KMedoidsFastPAM.Instance

Enclosing class:

KMedoidsFastPAM1<V>

protected static class KMedoidsFastPAM1.Instance
extends KMedoidsPAM.Instance

Instance for a single dataset.

Author:

Erich Schubert

Field Summary

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

assignment, distQ, ids, nearest, second

Constructor Summary

Constructors

Constructor and Description
Instance(DistanceQuery<?> distQ, DBIDs ids, WritableIntegerDataStore assignment) Constructor.

Method Summary

Modifier and Type	Method and Description
protected double	assignToNearestCluster(ArrayDBIDs means) Returns a list of clusters.
protected void	computeReassignmentCost(DBIDRef h, double[] cost) Compute the reassignment cost, for all medoids in one pass.
protected double	run(ArrayModifiableDBIDs medoids, int maxiter) Run the PAM optimization phase.
protected void	updateAssignment(ArrayModifiableDBIDs medoids, DBIDArrayIter miter, DBIDRef h, int m) Update an existing cluster assignment.
private int	updateSecondNearest(DBIDRef j, DBIDArrayIter medoids, int h, double dist_h, int n) Find the second nearest medoid.

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

computeReassignmentCost

Methods inherited from class java.lang.Object

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

Constructor Detail

Instance

public Instance(DistanceQuery<?> distQ,
                DBIDs ids,
                WritableIntegerDataStore assignment)

Constructor.

Parameters:

distQ - Distance query

ids - IDs to process

assignment - Cluster assignment

Method Detail

run

protected double run(ArrayModifiableDBIDs medoids,
                     int maxiter)

Run the PAM optimization phase.

Overrides:

run in class KMedoidsPAM.Instance

Parameters:

medoids - Medoids list

maxiter -

Returns:

cost

assignToNearestCluster

protected double assignToNearestCluster(ArrayDBIDs means)

Returns a list of clusters. The k^th cluster contains the ids of those objects, that are nearest to the k^th mean.

Overrides:

assignToNearestCluster in class KMedoidsPAM.Instance

Parameters:

means - Object centroids

Returns:

Assignment cost

computeReassignmentCost

protected void computeReassignmentCost(DBIDRef h,
                                       double[] cost)

Compute the reassignment cost, for all medoids in one pass.

Parameters:

h - Current object to swap with any medoid.

cost - Cost aggregation array, must have size k

updateAssignment

protected void updateAssignment(ArrayModifiableDBIDs medoids,
                                DBIDArrayIter miter,
                                DBIDRef h,
                                int m)

Update an existing cluster assignment.

Parameters:

medoids - Medoids set

miter - Medoid iterator

h - New medoid

m - Position of replaced medoid

updateSecondNearest

private int updateSecondNearest(DBIDRef j,
                                DBIDArrayIter medoids,
                                int h,
                                double dist_h,
                                int n)

Find the second nearest medoid.

Parameters:

j - Query point

medoids - Medoids

h - Known medoid

dist_h - Distance to h

n - Known nearest

Returns:

Index of second nearest medoid, KMedoidsPAM.Instance.second is updated.