keel.Algorithms.Semi_Supervised_Learning.Rel_RASCO

Class Rel_RASCOGenerator

java.lang.Object

keel.Algorithms.Semi_Supervised_Learning.Basic.PrototypeGenerator

keel.Algorithms.Semi_Supervised_Learning.Rel_RASCO.Rel_RASCOGenerator

public class Rel_RASCOGenerator
extends PrototypeGenerator

This class implements the Rel_RASCO algorithm. You can use: Knn, C4.5, SMO as classifiers.

Author:

triguero

Field Summary

Fields

Modifier and Type	Field and Description
protected int	numberOfClass Number of classes.
protected int	numberOfPrototypes Number of prototypes.

Fields inherited from class keel.Algorithms.Semi_Supervised_Learning.Basic.PrototypeGenerator

algorithmName, generatedDataSet, Instancetest, Instancetrain, SEED, seedDefaultValueList, testDataSet, trainingDataSet, transductiveDataSet

Constructor Summary

Constructors

Constructor and Description
Rel_RASCOGenerator(PrototypeSet _trainingDataSet, int neigbors, int poblacion, int perc, int iteraciones, double c1, double c2, double vmax, double wstart, double wend) Build a new Rel_RASCOGenerator Algorithm
Rel_RASCOGenerator(PrototypeSet t, PrototypeSet unlabeled, PrototypeSet test, Parameters parameters) Build a new Rel_RASCOGenerator Algorithm

Method Summary

Modifier and Type	Method and Description
Pair<PrototypeSet,PrototypeSet>	applyAlgorithm() Apply the Rel_RASCOGenerator method.
void	cambiarContextoAttributes() Changes the context of the attributes.
PrototypeSet	discretize(PrototypeSet label) It return a discretized prototypeset in 10 equal sized bins placed between the minimum and the maximum values
void	getSolicitaGarbageColector() Asks for the garbage collector.
static void	main(java.lang.String[] args) General main for all the prototoype generators Arguments: 0: Filename with the training data set to be condensed. 1: Filename which contains the test data set. 3: Seed of the random number generator.
double	probability(PrototypeSet label, int feature, int instance, boolean cond) Computes the reduction probability of a set for a certain value of the given instance and feature.
double[]	relevancia(PrototypeSet label) This methods return the mutual information between features and labels of the dataset.

Methods inherited from class keel.Algorithms.Semi_Supervised_Learning.Basic.PrototypeGenerator

absoluteAccuracy, absoluteAccuracyAndError, absoluteAccuracyKNN, accuracy, accuracy2, desordenar_vector_sin, desordenar_vector, execute, generateReducedDataSet, getResultingAccuracy, getResultsOfAccuracy, getSeed, getSetSizeFromPercentage, getSetSizeFromPercentage, getTime, inic_vector_sin, inic_vector, saveResultsOfAccuracyIn, selecRandomSet, setInstanceTest, setInstanceTrain, setSeed

Methods inherited from class java.lang.Object

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

Field Detail

numberOfPrototypes

protected int numberOfPrototypes

Number of prototypes.

numberOfClass

protected int numberOfClass

Number of classes.

Constructor Detail

Rel_RASCOGenerator

public Rel_RASCOGenerator(PrototypeSet _trainingDataSet,
                          int neigbors,
                          int poblacion,
                          int perc,
                          int iteraciones,
                          double c1,
                          double c2,
                          double vmax,
                          double wstart,
                          double wend)

Build a new Rel_RASCOGenerator Algorithm

Parameters:

_trainingDataSet - Original prototype set to be reduced.

neigbors - number of neighbours considered. (not used)

poblacion - population size. (not used)

perc - Reduction percentage of the prototype set.

iteraciones - number of iterations. (not used)

wend - ending w value. (not used)

c1 - class 1 value. (not used)

vmax - maximum v value. (not used)

c2 - class 2 value. (not used)

wstart - starting w value. (not used)

Rel_RASCOGenerator

public Rel_RASCOGenerator(PrototypeSet t,
                          PrototypeSet unlabeled,
                          PrototypeSet test,
                          Parameters parameters)

Build a new Rel_RASCOGenerator Algorithm

Parameters:

t - Original prototype set to be reduced.

unlabeled - Original unlabeled prototype set for SSL.

test - Origital test prototype set.

parameters - Parameters of the algorithm (only % of reduced set).

Method Detail

cambiarContextoAttributes

public void cambiarContextoAttributes()
                               throws java.lang.Exception

Changes the context of the attributes.

Throws:

java.lang.Exception - if the context can not be changed.

discretize

public PrototypeSet discretize(PrototypeSet label)

It return a discretized prototypeset in 10 equal sized bins placed between the minimum and the maximum values

Parameters:

label -

Returns:

probability

public double probability(PrototypeSet label,
                          int feature,
                          int instance,
                          boolean cond)

Computes the reduction probability of a set for a certain value of the given instance and feature.

Parameters:

label - prototype set.

feature - given feature.

instance - given instance.

cond - condicionated to be from the same class as the instance given.

Returns:

the reduction probability

relevancia

public double[] relevancia(PrototypeSet label)

This methods return the mutual information between features and labels of the dataset.

Parameters:

label -

Returns:

getSolicitaGarbageColector

public void getSolicitaGarbageColector()

Asks for the garbage collector.

applyAlgorithm

public Pair<PrototypeSet,PrototypeSet> applyAlgorithm()
                                               throws java.lang.Exception

Apply the Rel_RASCOGenerator method.

Overrides:

applyAlgorithm in class PrototypeGenerator

Returns:

transductive test sets.

Throws:

java.lang.Exception - if the algorithm can not be applied.

main

public static void main(java.lang.String[] args)

General main for all the prototoype generators Arguments: 0: Filename with the training data set to be condensed. 1: Filename which contains the test data set. 3: Seed of the random number generator. Always. **************************

Parameters:

args - Arguments of the main function.