keel.Algorithms.Hyperrectangles.EACH

Class RuleQualityEvaluation

java.lang.Object

keel.Algorithms.Hyperrectangles.EACH.RuleQualityEvaluation

public class RuleQualityEvaluation
extends java.lang.Object

To evaluate the rules

Since:

JDK1.2

Version:

1.1

Author:

Written by Rosa Venzala (University of Granada) 02/06/2008, Modified by Xavi Solé (La Salle, Ramón Llull University - Barcelona) 16/12/2008

Constructor Summary

Constructors

Constructor and Description
RuleQualityEvaluation(RuleSet conjreg, EachDataSet conjTrn, EachDataSet conjTst, int[] muestPorClaseTrain, int[] muestPorClaseTest, java.lang.String[] valorNombreClases, java.lang.String[] valorNombreClasesTest) Calculates the final statistical for a set of rules and a set of data

Method Summary

Modifier and Type	Method and Description
double	getAccuracyTest() Get Accuracyy Test
double	getAccuracyTrain() Get Accuracyy Train
java.lang.String	out(EachDataSet datos, boolean train) Generates a string with out-put lists
java.lang.String	printString() Prints on a string the statistical results

Methods inherited from class java.lang.Object

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

Constructor Detail

RuleQualityEvaluation

public RuleQualityEvaluation(RuleSet conjreg,
                             EachDataSet conjTrn,
                             EachDataSet conjTst,
                             int[] muestPorClaseTrain,
                             int[] muestPorClaseTest,
                             java.lang.String[] valorNombreClases,
                             java.lang.String[] valorNombreClasesTest)

Calculates the final statistical for a set of rules and a set of data

Parameters:

conjreg - Set of rules(complex) final

conjTrn - Set of data Train

conjTst - Set of data Test

muestPorClaseTrain - int[] Number of examples of each class in the Train set

muestPorClaseTest - int[] Number of examples of each class in the Test set

valorNombreClases - String[] Labels for each class in the Train set

valorNombreClasesTest - String[] Labels for each class in the test set

Method Detail

getAccuracyTrain

public double getAccuracyTrain()

Get Accuracyy Train

Returns:

percent

getAccuracyTest

public double getAccuracyTest()

Get Accuracyy Test

Returns:

percent

printString

public java.lang.String printString()

Prints on a string the statistical results

Returns:

a string with the statistical results

out

public java.lang.String out(EachDataSet datos,
                            boolean train)

Generates a string with out-put lists

Parameters:

datos - Set of data to compare with the set of rules

train - true if the data is from training dataset.

Returns:

A string with pairs (original class; calculated class;)