smile.clustering

Class CentroidClustering<T,U>

java.lang.Object

smile.clustering.PartitionClustering

smile.clustering.CentroidClustering<T,U>

Type Parameters:

T - the type of centroids.

U - the tpe of observations. Usually, T and U are the same. But in case of SIB, they are different.

All Implemented Interfaces:

java.io.Serializable, java.lang.Comparable<CentroidClustering<T,U>>

Direct Known Subclasses:

CLARANS, DeterministicAnnealing, GMeans, KMeans, KModes, SIB, XMeans

public abstract class CentroidClustering<T,U>
extends PartitionClustering
implements java.lang.Comparable<CentroidClustering<T,U>>

In centroid-based clustering, clusters are represented by a central vector, which may not necessarily be a member of the data set. When the number of clusters is fixed to k, k-means clustering gives a formal definition as an optimization problem: find the k cluster centers and assign the objects to the nearest cluster center, such that the squared distances from the cluster are minimized.

Variations of k-means include restricting the centroids to members of the data set (k-medoids), choosing medians (k-medians clustering), choosing the initial centers less randomly (k-means++) or allowing a fuzzy cluster assignment (fuzzy c-means), etc.

Most k-means-type algorithms require the number of clusters to be specified in advance, which is considered to be one of the biggest drawbacks of these algorithms. Furthermore, the algorithms prefer clusters of approximately similar size, as they will always assign an object to the nearest centroid. This often leads to incorrectly cut borders of clusters (which is not surprising since the algorithm optimizes cluster centers, not cluster borders).

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
T[]	centroids The centroids of each cluster.
double	distortion The total distortion.

Fields inherited from class smile.clustering.PartitionClustering

k, OUTLIER, size, y

Constructor Summary

Constructors

Constructor and Description
CentroidClustering(double distortion, T[] centroids, int[] y) Constructor.

Method Summary

Modifier and Type	Method and Description
int	compareTo(CentroidClustering<T,U> o)
abstract double	distance(T x, U y) The distance function.
int	predict(U x) Classifies a new observation.
java.lang.String	toString()

Methods inherited from class smile.clustering.PartitionClustering

run, seed

Methods inherited from class java.lang.Object

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

Field Detail

distortion

public final double distortion

The total distortion.

centroids

public final T[] centroids

The centroids of each cluster.

Constructor Detail

CentroidClustering

public CentroidClustering(double distortion,
                          T[] centroids,
                          int[] y)

Constructor.

Parameters:

distortion - the total distortion.

centroids - the centroids of each cluster.

y - the cluster labels.

Method Detail

compareTo

public int compareTo(CentroidClustering<T,U> o)

Specified by:

compareTo in interface java.lang.Comparable<CentroidClustering<T,U>>

distance

public abstract double distance(T x,
                                U y)

The distance function.

predict

public int predict(U x)

Classifies a new observation.

Parameters:

x - a new observation.

Returns:

the cluster label.

toString

public java.lang.String toString()

Overrides:

toString in class PartitionClustering