Package unifeat.featureSelection

Class FeatureSelection

java.lang.Object

unifeat.featureSelection.FeatureSelection

Direct Known Subclasses:

EmbeddedApproach, FeatureWeighting, FilterApproach, HybridApproach, WrapperApproach

public abstract class FeatureSelection
extends Object

The abstract class contains the main methods and fields that are used in all feature selection methods.

Author:

Sina Tabakhi

Field Summary

Fields

Modifier and Type	Field	Description
protected int	numClass
protected int	numFeatures
protected int	numSelectedFeature
protected int[]	selectedFeatureSubset
protected double[][]	trainSet

Constructor Summary

Constructors

Constructor	Description
FeatureSelection()	Initializes the parameters

Method Summary

Modifier and Type	Method	Description
abstract void	evaluateFeatures()	Starts the feature selection process by a given method
int[]	getSelectedFeatureSubset()	This method returns the subset of selected features by a given feature selection method.
void	loadDataSet(double[][] data, int numFeat, int numClasses)	Loads the dataset
void	loadDataSet(DatasetInfo ob)	Loads the dataset
void	setNumSelectedFeature(int numSelectedFeature)	This method sets the number of features that should be selected by a given feature selection method.
String	validate()	This method returns the potential errors in the input parameters.

Methods inherited from class java.lang.Object

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

Field Details

trainSet

protected double[][] trainSet

numFeatures

protected int numFeatures

numClass

protected int numClass

selectedFeatureSubset

protected int[] selectedFeatureSubset

numSelectedFeature

protected int numSelectedFeature

Constructor Details

FeatureSelection

public FeatureSelection()

Initializes the parameters

Method Details

loadDataSet

public void loadDataSet(DatasetInfo ob)

Loads the dataset

Parameters:

ob - an object of the DatasetInfo class

loadDataSet

public void loadDataSet(double[][] data,
 int numFeat,
 int numClasses)

Loads the dataset

Parameters:

data - the input dataset values

numFeat - the number of features in the dataset

numClasses - the number of classes in the dataset

evaluateFeatures

public abstract void evaluateFeatures()

Starts the feature selection process by a given method

validate

public String validate()

This method returns the potential errors in the input parameters.

Returns:

a string contains the information about incorrect parameters

setNumSelectedFeature

public void setNumSelectedFeature(int numSelectedFeature)

This method sets the number of features that should be selected by a given feature selection method.

Parameters:

numSelectedFeature - the number of selected features

getSelectedFeatureSubset

public int[] getSelectedFeatureSubset()

This method returns the subset of selected features by a given feature selection method.

Returns:

an array of subset of selected features