jsat.clustering.hierarchical

Class NNChainHAC

java.lang.Object

jsat.clustering.hierarchical.NNChainHAC

All Implemented Interfaces:

java.io.Serializable, Clusterer, KClusterer

public class NNChainHAC
extends java.lang.Object
implements KClusterer

This class implements Hierarchical Agglomerative Clustering via the Nearest Neighbor Chain approach. This runs in O(n²) time for any Lance Williams dissimilarity and uses O(n) memory.
This implementation also supports multi-threaded execution. see:

• Müllner, D. (2011). Modern hierarchical, agglomerative clustering algorithms. arXiv Preprint arXiv:1109.2378. Retrieved from here
• Murtagh, F., & Contreras, P. (2011). Methods of Hierarchical Clustering. In Data Mining and Knowledge Discovery. Wiley-Interscience.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
NNChainHAC() Creates a new NNChainHAC using the Ward method.
NNChainHAC(LanceWilliamsDissimilarity distMeasure) Creates a new NNChainHAC
NNChainHAC(LanceWilliamsDissimilarity distMeasure, DistanceMetric distance) Creates a new NNChain using the given dissimilarity measure and distance metric.

Method Summary

Modifier and Type	Method and Description
NNChainHAC	clone()
int[]	cluster(DataSet dataSet, boolean parallel, int[] designations) Performs clustering on the given data set.
int[]	cluster(DataSet dataSet, int clusters, boolean parallel, int[] designations)
int[]	cluster(DataSet dataSet, int lowK, int highK, boolean parallel, int[] designations)
int[]	getClusterDesignations(int[] designations, int clusters) Returns the assignment array for that would have been computed for the previous data set with the desired number of clusters.
java.util.List<java.util.List<DataPoint>>	getClusterDesignations(int clusters, DataSet data) Returns the assignment array for that would have been computed for the previous data set with the desired number of clusters.

Methods inherited from class java.lang.Object

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

Methods inherited from interface jsat.clustering.KClusterer

cluster, cluster, cluster, cluster, cluster, cluster

Methods inherited from interface jsat.clustering.Clusterer

cluster, cluster, cluster, supportsWeightedData

Constructor Detail

NNChainHAC

public NNChainHAC()

Creates a new NNChainHAC using the Ward method.

NNChainHAC

public NNChainHAC(LanceWilliamsDissimilarity distMeasure)

Creates a new NNChainHAC

Parameters:

distMeasure - the dissimilarity measure to use

NNChainHAC

public NNChainHAC(LanceWilliamsDissimilarity distMeasure,
                  DistanceMetric distance)

Creates a new NNChain using the given dissimilarity measure and distance metric. The correctness guarantees may not hold for distances other than the Euclidean distance, which is the norm for Hierarchical Cluster.

Parameters:

distMeasure - the dissimilarity measure to use

distance - the distance metric to use

Method Detail

clone

public NNChainHAC clone()

Specified by:

clone in interface Clusterer

Specified by:

clone in interface KClusterer

Overrides:

clone in class java.lang.Object

cluster

public int[] cluster(DataSet dataSet,
                     boolean parallel,
                     int[] designations)

Description copied from interface: Clusterer

Performs clustering on the given data set. Parameters may be estimated by the method, or other heuristics performed.

Specified by:

cluster in interface Clusterer

Parameters:

dataSet - the data set to perform clustering on

parallel - true if multiple threads should be used to perform clustering. false if it should be done in a single threaded manner.

designations - the array which will contain the designated values. The array will be altered and returned by the function. If null is given, a new array will be created and returned.

Returns:

the int[]

getClusterDesignations

public int[] getClusterDesignations(int[] designations,
                                    int clusters)

Returns the assignment array for that would have been computed for the previous data set with the desired number of clusters.

Parameters:

designations - the array to store the assignments in

clusters - the number of clusters desired

Returns:

the original array passed in, or null if no data set has been clustered.

See Also:

#hasStoredClustering()

getClusterDesignations

public java.util.List<java.util.List<DataPoint>> getClusterDesignations(int clusters,
                                                                        DataSet data)

Returns the assignment array for that would have been computed for the previous data set with the desired number of clusters.

Parameters:

clusters - the number of clusters desired

data -

Returns:

the list of data points in each cluster, or null if no data set has been clustered.

See Also:

#hasStoredClustering()

cluster

public int[] cluster(DataSet dataSet,
                     int clusters,
                     boolean parallel,
                     int[] designations)

Specified by:

cluster in interface KClusterer

cluster

public int[] cluster(DataSet dataSet,
                     int lowK,
                     int highK,
                     boolean parallel,
                     int[] designations)

Specified by:

cluster in interface KClusterer