Package unifeat.featureSelection.filter.unsupervised

Class RSM

java.lang.Object

unifeat.featureSelection.FeatureSelection

unifeat.featureSelection.filter.FilterApproach

unifeat.featureSelection.filter.unsupervised.RSM

public class RSM
extends FilterApproach

This java class is used to implement the random subspace method(RSM) method.

Author:

Sina Tabakhi

See Also:

• FilterApproach
• FeatureSelection

Field Summary

Fields

Modifier and Type	Field	Description
private int[]	featureScore
private final MultivariateMethodType	NAME_MULTI_APPROACH
private final int	NUM_ITERATION
private final int	SIZE_SUB_SPACE
private final int	THRESHOLD_ELIMINATION

Fields inherited from class unifeat.featureSelection.FeatureSelection

numClass, numFeatures, numSelectedFeature, selectedFeatureSubset, trainSet

Constructor Summary

Constructors

Constructor	Description
RSM(int sizeSelectedFeatureSubset, int numIter, int size, int threshold, MultivariateMethodType nameApproach)	Initializes the parameters
RSM(Object... arguments)	Initializes the parameters

Method Summary

Modifier and Type	Method	Description
private double[][]	createNewDataset(int[] index)	Creates a new dataset based on the given indeces of the features
void	evaluateFeatures()	Starts the feature selection process by random subspace method(RSM)
private int[]	multivariateApproach(double[][] data)	Selects the top feature with size THRESHOLD_ELIMINATION by given method
private void	permutation(int[] indexFeat, int seed)	Permutes the index of features
String	validate()	This method returns the potential errors in the input parameters.

Methods inherited from class unifeat.featureSelection.filter.FilterApproach

newMethod

Methods inherited from class unifeat.featureSelection.FeatureSelection

getSelectedFeatureSubset, loadDataSet, loadDataSet, setNumSelectedFeature

Methods inherited from class java.lang.Object

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

Field Details

NUM_ITERATION

private final int NUM_ITERATION

SIZE_SUB_SPACE

private final int SIZE_SUB_SPACE

THRESHOLD_ELIMINATION

private final int THRESHOLD_ELIMINATION

NAME_MULTI_APPROACH

private final MultivariateMethodType NAME_MULTI_APPROACH

featureScore

private int[] featureScore

Constructor Details

RSM

public RSM(Object... arguments)

Initializes the parameters

Parameters:

arguments - array of parameters contains (sizeSelectedFeatureSubset, numIter, size, threshold, nameApproach) in which sizeSelectedFeatureSubset is the number of selected features, numIter is the number of iteration in the RSM method, size is the size of the subspace, threshold is the number of selected features in each subspace, and nameApproach is the name of the multivariate approach used in the RSM

RSM

public RSM(int sizeSelectedFeatureSubset,
 int numIter,
 int size,
 int threshold,
 MultivariateMethodType nameApproach)

Initializes the parameters

Parameters:

sizeSelectedFeatureSubset - the number of selected features

numIter - the number of iteration in the RSM method

size - the size of the subspace

threshold - the number of selected features in each subspace

nameApproach - the name of the multivariate approach used in the RSM

Method Details

permutation

private void permutation(int[] indexFeat,
 int seed)

Permutes the index of features

Parameters:

indexFeat - the array of the index of features

seed - determines the index of the seed

multivariateApproach

private int[] multivariateApproach(double[][] data)

Selects the top feature with size THRESHOLD_ELIMINATION by given method

Parameters:

data - the new dataset

Returns:

the top feature with size THRESHOLD_ELIMINATION

createNewDataset

private double[][] createNewDataset(int[] index)

Creates a new dataset based on the given indeces of the features

Parameters:

index - an array of the indeces of features

Returns:

a new dataset

evaluateFeatures

public void evaluateFeatures()

Starts the feature selection process by random subspace method(RSM)

Specified by:

evaluateFeatures in class FeatureSelection

validate

public String validate()

This method returns the potential errors in the input parameters.

Overrides:

validate in class FeatureSelection

Returns:

a string contains the information about incorrect parameters