keel.Algorithms.Associative_Classification.ClassifierCPAR

Class Rule

java.lang.Object

keel.Algorithms.Associative_Classification.ClassifierCPAR.Rule

All Implemented Interfaces:

java.lang.Comparable

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

Class to store a non-fuzzy rule, together with some necessary information to manage the CBA algorithm

Since:

JDK1.5

Version:

1.0

Author:

Written by Jesus Alcala (University of Granada) 09/02/2010

Constructor Summary

Constructors

Constructor and Description
Rule() Copy Constructor
Rule(DataBase dataBase) Parameters Constructor
Rule(myDataset train, DataBase dataBase, int clas) Parameters Constructor

Method Summary

Modifier and Type	Method and Description
void	calculateLaplace(myDataset train) Function to calculate the Laplace accuracy to our rule from a train dataset
Rule	clone() Clone function.
int	compareTo(java.lang.Object a) Function to compare objects of the Rule class.
int	getClas() It returns the output class of the rule
java.util.ArrayList<Literal>	getGain(double min_gain, PNArray A) It returns an array of literals whose gain is higher than a minimum threshold
double	getLaplace() It returns the Laplace accuracy of the rule
int	getnAnts() It returns the number of antecedents of the rule
double	matching(int[] example) Function to check if a given example matchs with the rule (the rule correctly classifies it)
void	setLabel(int pos, int label) It sets the label for a given position in the antecedent (for a given attribute)

Methods inherited from class java.lang.Object

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

Constructor Detail

Rule

public Rule()

Copy Constructor

Rule

public Rule(DataBase dataBase)

Parameters Constructor

Parameters:

dataBase - Set of training data which is necessary to generate a rule

Rule

public Rule(myDataset train,
            DataBase dataBase,
            int clas)

Parameters Constructor

Parameters:

train - myDataset Set of training data (not used)

dataBase - DataBase Set of training data which is necessary to generate a rule

clas - output class value for the new rule

Method Detail

clone

public Rule clone()

Clone function.

Overrides:

clone in class java.lang.Object

Returns:

A copy of the Rule object.

setLabel

public void setLabel(int pos,
                     int label)

It sets the label for a given position in the antecedent (for a given attribute)

Parameters:

pos - int Location of the attribute which we want to set the label

label - int New label value to set

matching

public double matching(int[] example)

Function to check if a given example matchs with the rule (the rule correctly classifies it)

Parameters:

example - int[] Example to be classified

Returns:

double 0.0 = doesn't match, >0.0 = does.

getClas

public int getClas()

It returns the output class of the rule

Returns:

int Output class of the rule

getnAnts

public int getnAnts()

It returns the number of antecedents of the rule

Returns:

int Number of antecedents of the rule

getLaplace

public double getLaplace()

It returns the Laplace accuracy of the rule

Returns:

double Laplace accuracy of the rule

getGain

public java.util.ArrayList<Literal> getGain(double min_gain,
                                            PNArray A)

It returns an array of literals whose gain is higher than a minimum threshold

Parameters:

min_gain - double Minimum gain threshold

A - PNArray Training dataset splitted in Possitive and Negative examples

Returns:

ArrayList Literals whose gain is higher than a minimum threshold

calculateLaplace

public void calculateLaplace(myDataset train)

Function to calculate the Laplace accuracy to our rule from a train dataset

Parameters:

train - myDataset Dataset used to calculate Laplace

compareTo

public int compareTo(java.lang.Object a)

Function to compare objects of the Rule class. Necessary to be able to use "sort" function It sorts in an decreasing order of laplace accuracy

Specified by:

compareTo in interface java.lang.Comparable

Parameters:

a - Rule object to compare with.

Returns:

1 if a is bigger, -1 if smaller and 0 otherwise.