weka.classifiers.trees.j48
Class C45Split

java.lang.Object
  weka.classifiers.trees.j48.ClassifierSplitModel
      weka.classifiers.trees.j48.C45Split

All Implemented Interfaces:

java.lang.Cloneable, java.io.Serializable

public class C45Split

extends ClassifierSplitModel

Class implementing a C4.5-type split on an attribute.

Version:

$Revision: 1.8 $

Author:

Eibe Frank (eibe@cs.waikato.ac.nz)

See Also:

Serialized Form

Field Summary

private static GainRatioSplitCrit	gainRatioCrit Static reference to splitting criterion.
private static InfoGainSplitCrit	infoGainCrit Static reference to splitting criterion.
private int	m_attIndex Attribute to split on.
private int	m_complexityIndex Desired number of branches.
private double	m_gainRatio GainRatio of split.
private int	m_index Number of split points.
private double	m_infoGain InfoGain of split.
private int	m_minNoObj Minimum number of objects in a split.
private double	m_splitPoint Value of split point.
private double	m_sumOfWeights The sum of the weights of the instances.

Fields inherited from class weka.classifiers.trees.j48.ClassifierSplitModel

m_distribution, m_numSubsets

Constructor Summary

C45Split(int attIndex, int minNoObj, double sumOfWeights)
Initializes the split model.

Method Summary

int	attIndex() Returns index of attribute for which split was generated.
void	buildClassifier(Instances trainInstances) Creates a C4.5-type split on the given data.
double	classProb(int classIndex, Instance instance, int theSubset) Gets class probability for instance.
double	codingCost() Returns coding cost for split (used in rule learner).
double	gainRatio() Returns (C4.5-type) gain ratio for the generated split.
private void	handleEnumeratedAttribute(Instances trainInstances) Creates split on enumerated attribute.
private void	handleNumericAttribute(Instances trainInstances) Creates split on numeric attribute.
double	infoGain() Returns (C4.5-type) information gain for the generated split.
java.lang.String	leftSide(Instances data) Prints left side of condition..
double[][]	minsAndMaxs(Instances data, double[][] minsAndMaxs, int index) Returns the minsAndMaxs of the index.th subset.
void	resetDistribution(Instances data) Sets distribution associated with model.
java.lang.String	rightSide(int index, Instances data) Prints the condition satisfied by instances in a subset.
void	setSplitPoint(Instances allInstances) Sets split point to greatest value in given data smaller or equal to old split point.
java.lang.String	sourceExpression(int index, Instances data) Returns a string containing java source code equivalent to the test made at this node.
double[]	weights(Instance instance) Returns weights if instance is assigned to more than one subset.
int	whichSubset(Instance instance) Returns index of subset instance is assigned to.

Methods inherited from class weka.classifiers.trees.j48.ClassifierSplitModel

checkModel, classifyInstance, classProbLaplace, clone, distribution, dumpLabel, dumpModel, numSubsets, sourceClass, split

Methods inherited from class java.lang.Object

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

Field Detail

m_complexityIndex

private int m_complexityIndex

Desired number of branches.

m_attIndex

private int m_attIndex

Attribute to split on.

m_minNoObj

private int m_minNoObj

Minimum number of objects in a split.

m_splitPoint

private double m_splitPoint

Value of split point.

m_infoGain

private double m_infoGain

InfoGain of split.

m_gainRatio

private double m_gainRatio

GainRatio of split.

m_sumOfWeights

private double m_sumOfWeights

The sum of the weights of the instances.

m_index

private int m_index

Number of split points.

infoGainCrit

private static InfoGainSplitCrit infoGainCrit

Static reference to splitting criterion.

gainRatioCrit

private static GainRatioSplitCrit gainRatioCrit

Static reference to splitting criterion.

Constructor Detail

C45Split

public C45Split(int attIndex,
                int minNoObj,
                double sumOfWeights)

Initializes the split model.

Method Detail

buildClassifier

public void buildClassifier(Instances trainInstances)
                     throws java.lang.Exception

Creates a C4.5-type split on the given data. Assumes that none of the class values is missing.

Specified by:

buildClassifier in class ClassifierSplitModel

Throws:

java.lang.Exception - if something goes wrong

attIndex

public final int attIndex()

Returns index of attribute for which split was generated.

classProb

public final double classProb(int classIndex,
                              Instance instance,
                              int theSubset)
                       throws java.lang.Exception

Gets class probability for instance.

Overrides:

classProb in class ClassifierSplitModel

Throws:

java.lang.Exception - if something goes wrong

codingCost

public final double codingCost()

Returns coding cost for split (used in rule learner).

Overrides:

codingCost in class ClassifierSplitModel

gainRatio

public final double gainRatio()

Returns (C4.5-type) gain ratio for the generated split.

handleEnumeratedAttribute

private void handleEnumeratedAttribute(Instances trainInstances)
                                throws java.lang.Exception

Creates split on enumerated attribute.

Throws:

java.lang.Exception - if something goes wrong

handleNumericAttribute

private void handleNumericAttribute(Instances trainInstances)
                             throws java.lang.Exception

Creates split on numeric attribute.

Throws:

java.lang.Exception - if something goes wrong

infoGain

public final double infoGain()

Returns (C4.5-type) information gain for the generated split.

leftSide

public final java.lang.String leftSide(Instances data)

Prints left side of condition..

Specified by:

leftSide in class ClassifierSplitModel

Parameters:

data - training set.

rightSide

public final java.lang.String rightSide(int index,
                                        Instances data)

Prints the condition satisfied by instances in a subset.

Specified by:

rightSide in class ClassifierSplitModel

Parameters:

index - of subset

data - training set.

sourceExpression

public final java.lang.String sourceExpression(int index,
                                               Instances data)

Returns a string containing java source code equivalent to the test made at this node. The instance being tested is called "i".

Specified by:

sourceExpression in class ClassifierSplitModel

Parameters:

index - index of the nominal value tested

data - the data containing instance structure info

Returns:

a value of type 'String'

setSplitPoint

public final void setSplitPoint(Instances allInstances)

Sets split point to greatest value in given data smaller or equal to old split point. (C4.5 does this for some strange reason).

minsAndMaxs

public final double[][] minsAndMaxs(Instances data,
                                    double[][] minsAndMaxs,
                                    int index)

Returns the minsAndMaxs of the index.th subset.

resetDistribution

public void resetDistribution(Instances data)
                       throws java.lang.Exception

Sets distribution associated with model.

Overrides:

resetDistribution in class ClassifierSplitModel

Throws:

java.lang.Exception

weights

public final double[] weights(Instance instance)

Returns weights if instance is assigned to more than one subset. Returns null if instance is only assigned to one subset.

Specified by:

weights in class ClassifierSplitModel

whichSubset

public final int whichSubset(Instance instance)
                      throws java.lang.Exception

Returns index of subset instance is assigned to. Returns -1 if instance is assigned to more than one subset.

Specified by:

whichSubset in class ClassifierSplitModel

Throws:

java.lang.Exception - if something goes wrong