keel.Algorithms.Subgroup_Discovery.aprioriSD

Class EvaluaCalidadReglas

java.lang.Object

keel.Algorithms.Subgroup_Discovery.aprioriSD.EvaluaCalidadReglas

public class EvaluaCalidadReglas
extends java.lang.Object

Get the statistical data from the algorithm. In particular, it evaluates the quality of the rules generated.

Since:

JDK1.2

Version:

1.1

Author:

Alberto Fernández Hilario

Constructor Summary

Constructors

Constructor and Description
EvaluaCalidadReglas(ConjReglas conjreg, ConjDatos conjTrn, ConjDatos conjTst, int[] muestPorClaseTrain, int[] muestPorClaseTest, java.lang.String[] valorNombreClases) Calculates the final statisticals for a set of rules and a set of data

Method Summary

Modifier and Type	Method and Description
void	ajustaDistribucion(ConjDatos train) Updates the dataset distribution.
java.lang.String	printString() Prints on a string the statistical(for test)
java.lang.String	salida(ConjDatos datos, int conj) Generates a string with the out-put lists

Methods inherited from class java.lang.Object

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

Constructor Detail

EvaluaCalidadReglas

public EvaluaCalidadReglas(ConjReglas conjreg,
                           ConjDatos conjTrn,
                           ConjDatos conjTst,
                           int[] muestPorClaseTrain,
                           int[] muestPorClaseTest,
                           java.lang.String[] valorNombreClases)

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

Parameters:

conjreg - Set of rules(complex) final

conjTrn - Set of train data

conjTst - Set of test data

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

muestPorClaseTest - int[] Number of examples of each class in train test

valorNombreClases - String[] Labels for each class

Method Detail

printString

public java.lang.String printString()

Prints on a string the statistical(for test)

Returns:

the statisticals

ajustaDistribucion

public void ajustaDistribucion(ConjDatos train)

Updates the dataset distribution.

Parameters:

train - given dataset.

salida

public java.lang.String salida(ConjDatos datos,
                               int conj)

Generates a string with the out-put lists

Parameters:

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

conj - flag to indicates if the set is for training or test (0,1).

Returns:

A list of pairs: 'original class;calculated class;'