edu.stanford.nlp.cluster
Class Kmeans

java.lang.Object
  |
  +--edu.stanford.nlp.cluster.AbstractClusteringMethod
        |
        +--edu.stanford.nlp.cluster.Kmeans

All Implemented Interfaces:

ClusteringMethod

public class Kmeans

extends AbstractClusteringMethod

K-means Clustering Algorithm.

Field Summary

Fields inherited from class edu.stanford.nlp.cluster.AbstractClusteringMethod

clusters, db, method, nc, nd, nt

Constructor Summary

Kmeans()

Method Summary

void	assignData() Assigns each data in data set to it closest mean
void	assignDatum(int index) Assigns a datum to its closest mean
SimpleClusters	cluster(DataCollection data, int num_clusters) Clusters by performing 30 iterations
void	evaluateMeans() Evaluates the means of each cluster
void	finish() Finalizes clusters by setting P(w|z) to be the weighted mean of the documents in the cluster
void	initialize() Initializes clusters by assigning each point randomly to a cluster
void	oneIteration() Performs one iteration: a) Evaluates cluster means, b) Assigns each datum to its closest mean

Methods inherited from class edu.stanford.nlp.cluster.AbstractClusteringMethod

cluster, evaluate, evaluate, initialize, toString, toXMLString

Methods inherited from class java.lang.Object

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

Constructor Detail

Kmeans

public Kmeans()

Method Detail

initialize

public void initialize()

Initializes clusters by assigning each point randomly to a cluster

assignDatum

public void assignDatum(int index)

Assigns a datum to its closest mean

assignData

public void assignData()

Assigns each data in data set to it closest mean

evaluateMeans

public void evaluateMeans()

Evaluates the means of each cluster

oneIteration

public void oneIteration()

Performs one iteration: a) Evaluates cluster means, b) Assigns each datum to its closest mean

cluster

public SimpleClusters cluster(DataCollection data,
                              int num_clusters)

Clusters by performing 30 iterations

finish

public void finish()

Finalizes clusters by setting P(w|z) to be the weighted mean of the documents in the cluster