keel.Algorithms.ImbalancedClassification.ImbalancedAlgorithms.GP_COACH_H

Class CHC

java.lang.Object

keel.Algorithms.ImbalancedClassification.ImbalancedAlgorithms.GP_COACH_H.CHC

public class CHC
extends java.lang.Object

Title: CHC

Description: Uses a CHC algorithm to select the rules used in the GP-COACH-H algorithm

Company: KEEL

Since:

JDK1.5

Version:

1.5

Author:

Written by Victoria Lopez (University of Granada) 26/04/2011

Constructor Summary

Constructors

Constructor and Description
CHC() Default constructor
CHC(myDataset current_dataset, GP_COACH_H gp_coach_data, double alpha, java.util.ArrayList<Rule> pop, int n_low, int n_high, int eval, int popLength, int bits_per_gene) Creates a CHC object with its parameters

Method Summary

Modifier and Type	Method and Description
double[]	obtainLateralTuning() Obtains the best lateral tuning from the genetic tuning process
java.util.ArrayList<Rule>	obtainNewRuleBase() Obtains the best set of rules from the genetic rule selection process
void	runCHC() Run the CHC algorithm for the data in this population

Methods inherited from class java.lang.Object

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

Constructor Detail

CHC

public CHC()

Default constructor

CHC

public CHC(myDataset current_dataset,
           GP_COACH_H gp_coach_data,
           double alpha,
           java.util.ArrayList<Rule> pop,
           int n_low,
           int n_high,
           int eval,
           int popLength,
           int bits_per_gene)

Creates a CHC object with its parameters

Parameters:

current_dataset - Training dataset used in this algorithm

gp_coach_data - GP-COACH-H object that we will use to train an element in the training set with our rules

alpha - Alpha of the raw_fitness evaluation function

pop - Population of rules we want to select

n_low - Number of low granularity rules in the rule base

n_high - Number of high granularity rules in the rule base

eval - Maximum number of evaluations in the CHC algorithm

popLength - Size of the population in the CHC algorithm

bits_per_gene - Bits per gene for the gray code associated to the real part of the CHC algorithm (lateral tuning)

Method Detail

runCHC

public void runCHC()

Run the CHC algorithm for the data in this population

obtainNewRuleBase

public java.util.ArrayList<Rule> obtainNewRuleBase()

Obtains the best set of rules from the genetic rule selection process

Returns:

best set of rules

obtainLateralTuning

public double[] obtainLateralTuning()

Obtains the best lateral tuning from the genetic tuning process

Returns:

real matrix representing the best lateral tuning that needs to be applied to the data base