Package unifeat.featureSelection.embedded.SVMBasedMethods

Class MSVM_RFE

java.lang.Object

unifeat.featureSelection.FeatureSelection

unifeat.featureSelection.embedded.EmbeddedApproach

unifeat.featureSelection.embedded.SVMBasedMethods.SVMBasedMethods

unifeat.featureSelection.embedded.SVMBasedMethods.MSVM_RFE

public class MSVM_RFE
extends SVMBasedMethods

This java class is used to implement MSVM_RFE method for binary classification based on SVM_RFE method (support vector machine method based on recursive feature elimination) in which multiple linear SVMs trained on subsamples of the original training data. K-fold cross validation is used as the resampling method.

Author:

Sina Tabakhi

See Also:

• SVMBasedMethods
• EmbeddedApproach
• FeatureSelection

Field Summary

Fields

Modifier and Type	Field	Description
private final double	ERROR_DENOMINATOR
private int	kFoldValue
private int	numRun

Fields inherited from class unifeat.featureSelection.embedded.SVMBasedMethods.SVMBasedMethods

classLabelInTrainSet, kernelType, parameterC, TEMP_PATH

Fields inherited from class unifeat.featureSelection.embedded.EmbeddedApproach

classLabel, nameFeatures, PROJECT_PATH

Fields inherited from class unifeat.featureSelection.FeatureSelection

numClass, numFeatures, numSelectedFeature, selectedFeatureSubset, trainSet

Constructor Summary

Constructors

Constructor	Description
MSVM_RFE(Object... arguments)	Initializes the parameters
MSVM_RFE(String path, SVMKernelType kernelType, double c, int kFoldValue, int numRun)	Initializes the parameters

Method Summary

Modifier and Type	Method	Description
protected double[][]	buildSVM_KFoldCrossValidation(int[] selectedFeature)	Generates binary classifiers (SVM by applying k-fold cross validation resampling strategy) using input data and based on selected feature subset.
void	evaluateFeatures()	Starts the feature selection process by multiple support vector machine method based on recursive feature elimination using k-fold cross validation resampling strategy (MSVM_RFE)
private double[]	getFeaturesWeights(int[] selectedFeature)	Generates binary classifiers (SVM by applying k-fold cross validation resampling strategy) using input data and based on selected feature subset, and finally returns the weights of features
String	validate()	This method returns the potential errors in the input parameters.

Methods inherited from class unifeat.featureSelection.embedded.SVMBasedMethods.SVMBasedMethods

buildSVM_OneAgainstOne, buildSVM_OneAgainstRest, createClassLabel

Methods inherited from class unifeat.featureSelection.embedded.EmbeddedApproach

loadDataSet, loadDataSet, newMethod, originalFeatureSet

Methods inherited from class unifeat.featureSelection.FeatureSelection

getSelectedFeatureSubset, setNumSelectedFeature

Methods inherited from class java.lang.Object

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

Field Details

ERROR_DENOMINATOR

private final double ERROR_DENOMINATOR

See Also:

• Constant Field Values

kFoldValue

private int kFoldValue

numRun

private int numRun

Constructor Details

MSVM_RFE

public MSVM_RFE(Object... arguments)

Initializes the parameters

Parameters:

arguments - array of parameters contains (path, kernelType, Parameter c, kFoldValue, numRun) in which path is the path of the project, kernelType is the type of kernel to use, Parameter c is the complexity parameter C, kFoldValue is the number of subsamples in k-fold cross validation, and numRun is the number of multiple runs of k-fold CV

MSVM_RFE

public MSVM_RFE(String path,
 SVMKernelType kernelType,
 double c,
 int kFoldValue,
 int numRun)

Initializes the parameters

Parameters:

path - the path of the project

kernelType - the type of kernel to use

c - the complexity parameter C

kFoldValue - the number of subsamples in k-fold cross validation

numRun - the number of multiple runs of k-fold CV

Method Details

buildSVM_KFoldCrossValidation

protected double[][] buildSVM_KFoldCrossValidation(int[] selectedFeature)

Generates binary classifiers (SVM by applying k-fold cross validation resampling strategy) using input data and based on selected feature subset.

Parameters:

selectedFeature - an array of indices of the selected feature subset

Returns:

an array of the weights of features

getFeaturesWeights

private double[] getFeaturesWeights(int[] selectedFeature)

Generates binary classifiers (SVM by applying k-fold cross validation resampling strategy) using input data and based on selected feature subset, and finally returns the weights of features

Parameters:

selectedFeature - an array of indices of the selected feature subset

Returns:

an array of the weights of features

evaluateFeatures

public void evaluateFeatures()

Starts the feature selection process by multiple support vector machine method based on recursive feature elimination using k-fold cross validation resampling strategy (MSVM_RFE)

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