smile.imputation

Class KMeansImputation

java.lang.Object

smile.imputation.KMeansImputation

All Implemented Interfaces:

MissingValueImputation

public class KMeansImputation
extends Object
implements MissingValueImputation

Missing value imputation by K-Means clustering. First cluster data by KMeans and then impute missing values with the average value of each attribute in the clusters.

Author:

Haifeng Li

Constructor Summary

Constructors

Constructor and Description
KMeansImputation(int k) Constructor.
KMeansImputation(int k, int runs) Constructor.

Method Summary

Methods

Modifier and Type	Method and Description
void	impute(double[][] data) Impute missing values in the dataset.

Methods inherited from class java.lang.Object

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

Constructor Detail

KMeansImputation

public KMeansImputation(int k)

Constructor.

Parameters:

k - the number of clusters in K-Means clustering.

KMeansImputation

public KMeansImputation(int k,
                int runs)

Constructor.

Parameters:

k - the number of clusters in K-Means clustering.

runs - the number of runs of K-Means algorithm.

Method Detail

impute

public void impute(double[][] data)
            throws MissingValueImputationException

Description copied from interface: MissingValueImputation

Impute missing values in the dataset.

Specified by:

impute in interface MissingValueImputation

Parameters:

data - a data set with missing values (represented as Double.NaN). On output, missing values are filled with estimated values.

Throws:

MissingValueImputationException