Package adaa.analytics.rules.logic.induction

Class ClassificationSnC

java.lang.Object

adaa.analytics.rules.logic.induction.AbstractSeparateAndConquer

adaa.analytics.rules.logic.induction.ClassificationSnC

Direct Known Subclasses:

ApproximateClassificationSnC, ClassificationExpertSnC, ContrastSnC

public class ClassificationSnC
extends AbstractSeparateAndConquer

Separate'n'conquer algorithm for generating classification rule sets.

Field Summary

Fields

Modifier and Type	Field	Description
protected AbstractFinder	finder	Module for finding single classification rules.

Fields inherited from class adaa.analytics.rules.logic.induction.AbstractSeparateAndConquer

factory, operatorCommandProxy, params, setFactory

Constructor Summary

Constructors

Constructor	Description
ClassificationSnC(AbstractFinder finder, InductionParameters params)

Method Summary

Modifier and Type	Method	Description
void	preprocessClass(IExampleSet dataset, int classId)
RuleSetBase	run(IExampleSet dataset)	Generates a classification rule set on the basis of a training set.

Methods inherited from class adaa.analytics.rules.logic.induction.AbstractSeparateAndConquer

setOperatorCommandProxy

Methods inherited from class java.lang.Object

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

Field Details

finder

protected AbstractFinder finder

Module for finding single classification rules.

Constructor Details

ClassificationSnC

public ClassificationSnC(AbstractFinder finder,
 InductionParameters params)

Method Details

run

public RuleSetBase run(IExampleSet dataset)

Generates a classification rule set on the basis of a training set.

Specified by:

run in class AbstractSeparateAndConquer

Parameters:

dataset - Training data set.

Returns:

Rule set.

preprocessClass

public void preprocessClass(IExampleSet dataset,
 int classId)