keel.Algorithms.Preprocess.Feature_Selection.Shared

Class FSAlgorithm

java.lang.Object

keel.Algorithms.Preprocess.Feature_Selection.Shared.FSAlgorithm

Direct Known Subclasses:

GGA

public abstract class FSAlgorithm
extends java.lang.Object

File: FSAlgorithm.java A general framework for FS Algorithms. This class contains all common operations in the developement of a FS algorithm.

Since:

JDK1.5

Version:

1.0

Author:

Written by Joaquín Derrac (University of Granada) 13/11/2008

Field Summary

Fields

Modifier and Type	Field and Description
protected long	initialTime Initial time.
protected int	inputAtt Number of input attributes
protected Attribute[]	inputs Inputs attributes
protected java.lang.String	name Naming.
protected int	nClasses Number of classes.
protected int[]	nInstances Number of instances of each classes.
protected boolean[]	nulls Missing values of a instance
protected java.lang.String[]	outFile Output files names
protected Attribute	output Output attribute
protected java.lang.String	referenceFile Reference file name
protected java.lang.String	relation Relation string.
protected long	seed Random seed.
protected Instance	temp Temporal instance.
protected InstanceSet	test Test dataset
protected double[][]	testData Test input data.
protected java.lang.String	testFile Test file name
protected int[]	testOutput Test output data.
protected InstanceSet	train Training dataset
protected double[][]	trainData Training input data.
protected java.lang.String	trainFile Train file name
protected int[][]	trainNominal Training nominal input data.
protected boolean[][]	trainNulls Training missing input data.
protected int[]	trainOutput Training output data.
protected double[][]	trainReal Training real input data.

Constructor Summary

Constructors

Constructor and Description
FSAlgorithm()

Method Summary

Modifier and Type	Method and Description
protected double	euclideanDistance(double[] instance1, double[] instance2) Calculates the Euclidean distance between two instances
protected void	normalize() This function builds the data matrix for training data and normalizes inputs values
static java.lang.String	printInstance(int[] instance) Generates a string with the contents of the instance
protected void	readConfiguracion(java.lang.String script) Reads configuration script, and extracts its contents.
protected void	readDataFiles(java.lang.String script) Read the configuration and data files, and process it.
protected abstract void	readParameters(java.lang.String script) Reads the parameters of the algorithm.
protected boolean	same(double[] a, double[] b) Checks if two instances are the same
protected void	setInitialTime() Sets the time counter

Methods inherited from class java.lang.Object

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

Field Detail

outFile

protected java.lang.String[] outFile

Output files names

testFile

protected java.lang.String testFile

Test file name

trainFile

protected java.lang.String trainFile

Train file name

referenceFile

protected java.lang.String referenceFile

Reference file name

train

protected InstanceSet train

Training dataset

test

protected InstanceSet test

Test dataset

temp

protected Instance temp

Temporal instance.

inputAtt

protected int inputAtt

Number of input attributes

inputs

protected Attribute[] inputs

Inputs attributes

output

protected Attribute output

Output attribute

nulls

protected boolean[] nulls

Missing values of a instance

trainData

protected double[][] trainData

Training input data.

trainReal

protected double[][] trainReal

Training real input data.

trainOutput

protected int[] trainOutput

Training output data.

trainNominal

protected int[][] trainNominal

Training nominal input data.

trainNulls

protected boolean[][] trainNulls

Training missing input data.

testData

protected double[][] testData

Test input data.

testOutput

protected int[] testOutput

Test output data.

relation

protected java.lang.String relation

Relation string.

nClasses

protected int nClasses

Number of classes.

nInstances

protected int[] nInstances

Number of instances of each classes.

initialTime

protected long initialTime

Initial time.

name

protected java.lang.String name

Naming.

seed

protected long seed

Random seed.

Constructor Detail

FSAlgorithm

public FSAlgorithm()

Method Detail

readDataFiles

protected void readDataFiles(java.lang.String script)

Read the configuration and data files, and process it.

Parameters:

script - Name of the configuration script

readConfiguracion

protected void readConfiguracion(java.lang.String script)

Reads configuration script, and extracts its contents.

Parameters:

script - Name of the configuration script

readParameters

protected abstract void readParameters(java.lang.String script)

Reads the parameters of the algorithm. Must be implemented in the subclass.

Parameters:

script - Configuration script

normalize

protected void normalize()
                  throws CheckException

This function builds the data matrix for training data and normalizes inputs values

Throws:

CheckException - if the dataset is not appropiate for this algorithm.

euclideanDistance

protected double euclideanDistance(double[] instance1,
                                   double[] instance2)

Calculates the Euclidean distance between two instances

Parameters:

instance1 - First instance

instance2 - Second instance

Returns:

The Euclidean distance

same

protected boolean same(double[] a,
                       double[] b)

Checks if two instances are the same

Parameters:

a - First instance

b - Second instance

Returns:

True if both instances are equal.

printInstance

public static java.lang.String printInstance(int[] instance)

Generates a string with the contents of the instance

Parameters:

instance - Instance to print.

Returns:

instance A string, with the values of the instance

setInitialTime

protected void setInitialTime()

Sets the time counter