keel.Algorithms.Fuzzy_Rule_Learning.Genetic.ClassifierFuzzySGERD

Class Rule

java.lang.Object

keel.Algorithms.Fuzzy_Rule_Learning.Genetic.ClassifierFuzzySGERD.Rule

All Implemented Interfaces:

java.lang.Comparable

public class Rule
extends java.lang.Object
implements java.lang.Comparable

Codifies a Fuzzy Rule

Since:

JDK1.5

Version:

1.5

Author:

Written by Alberto Fernández (University of Granada) 29/10/2007, Modified by Jesus Alcala Fernandez (University of Granada) 20/05/2009

Constructor Summary

Constructors

Constructor and Description
Rule(DataBase dataBase, int typeEvaluation) Constructor with parameters
Rule(Rule r) Copy Constructor

Method Summary

Modifier and Type	Method and Description
void	asignAntecedent(int[] antecedent) It assings the ancedent of the Rule
Rule	clone() It carries out a copy of the current rule
int	compareTo(java.lang.Object a) Compares the fitness of two rules for the ordering procedure
double	compatibility(double[] example) It computes the compatibility degree of an example for this rule
void	computeConsequent(myDataset train) Procedure to compute the best consequent for a given rule
double	evaluation(myDataset train) Evaluates the rule for computing its fitness
int	getActive() It returns the activation of this rule
int	getClas() It returns the class of the rule
int	getPosActive(int value) It returns the number of valid labels for the rule until a given position
boolean	isActive(int pos) It returns if a given condition is active or not
boolean	isNew() It returns if the rule is new or not
int	lengthRule() It returns the length of the rule
void	offNew() Sets this rule as "non-new"
void	onNew() Sets this rule as new
void	setClass(int clas) Sets the class for the rule
void	setConsequent(myDataset train) It sets the class for this rule
void	setLabel(int pos, int label) It sets a new label for the rule

Methods inherited from class java.lang.Object

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

Constructor Detail

Rule

public Rule(Rule r)

Copy Constructor

Parameters:

r - Rule the rule to be copied

Rule

public Rule(DataBase dataBase,
            int typeEvaluation)

Constructor with parameters

Parameters:

dataBase - DataBase the DB

typeEvaluation - int a code for the compatibility degre computation

Method Detail

asignAntecedent

public void asignAntecedent(int[] antecedent)

It assings the ancedent of the Rule

Parameters:

antecedent - int[] An array containing the fuzzy labels

clone

public Rule clone()

It carries out a copy of the current rule

Overrides:

clone in class java.lang.Object

Returns:

Rule an exact copy of the rule

computeConsequent

public void computeConsequent(myDataset train)

Procedure to compute the best consequent for a given rule

Parameters:

train - myDataset the tranning set

evaluation

public double evaluation(myDataset train)

Evaluates the rule for computing its fitness

Parameters:

train - myDataset the trianing set

Returns:

double the fitness of the rule

setClass

public void setClass(int clas)

Sets the class for the rule

Parameters:

clas - int the class id

setConsequent

public void setConsequent(myDataset train)

It sets the class for this rule

Parameters:

train - myDataset the training set

compatibility

public double compatibility(double[] example)

It computes the compatibility degree of an example for this rule

Parameters:

example - double[] the example

Returns:

double the compatibility degree

onNew

public void onNew()

Sets this rule as new

offNew

public void offNew()

Sets this rule as "non-new"

isNew

public boolean isNew()

It returns if the rule is new or not

Returns:

boolean true if the rule is new, false otherwise

getClas

public int getClas()

It returns the class of the rule

Returns:

int the class of the rule

getActive

public int getActive()

It returns the activation of this rule

Returns:

int the activation of this rule

getPosActive

public int getPosActive(int value)

It returns the number of valid labels for the rule until a given position

Parameters:

value - int the position of the rule

Returns:

int the number of valid labels for the rule until a given position

lengthRule

public int lengthRule()

It returns the length of the rule

Returns:

int the length of the rule

isActive

public boolean isActive(int pos)

It returns if a given condition is active or not

Parameters:

pos - int the position

Returns:

boolean true if the condition has a valid label, false otherwise

setLabel

public void setLabel(int pos,
                     int label)

It sets a new label for the rule

Parameters:

pos - int the position in the antecedent

label - int the label id

compareTo

public int compareTo(java.lang.Object a)

Compares the fitness of two rules for the ordering procedure

Specified by:

compareTo in interface java.lang.Comparable

Parameters:

a - Object a Rule

Returns:

int -1 if the current rule is worst than the one that is compared, 1 for the contrary case and 0 if they are equal.