All Packages  Class Hierarchy  This Package  Previous  Next  Index  WEKA's home

Class weka.classifiers.j48.C45Split

java.lang.Object
   |
   +----weka.classifiers.j48.ClassifierSplitModel
           |
           +----weka.classifiers.j48.C45Split

public class C45Split

extends ClassifierSplitModel

Version:

$Revision: 1.3 $

Author:

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

C45Split(int, int, double)

Initializes the split model.

attIndex()

Returns index of attribute for which split was generated.

buildClassifier(Instances)

Creates a C4.5-type split on the given data.

classProb(int, Instance)

Gets class probability for instance.

codingCost()

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

gainRatio()

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

infoGain()

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

leftSide(Instances)

Prints left side of condition..

minsAndMaxs(Instances, double[][], int)

Returns the minsAndMaxs of the index.th subset.

rightSide(int, Instances)

Prints the condition satisfied by instances in a subset.

setSplitPoint(Instances)

Sets split point to greatest value in given data smaller or equal to old split point.

sourceExpression(int, Instances)

Returns a string containing java source code equivalent to the test made at this node.

weights(Instance)

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

whichSubset(Instance)

Returns index of subset instance is assigned to.

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

Initializes the split model.

 public void buildClassifier(Instances trainInstances) throws Exception

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

Throws: Exception

if something goes wrong

Overrides:

buildClassifier in class ClassifierSplitModel

 public final int attIndex()

Returns index of attribute for which split was generated.

 public final double classProb(int classIndex,
                               Instance instance) throws Exception

Gets class probability for instance.

Throws: Exception

if something goes wrong

Overrides:

classProb in class ClassifierSplitModel

 public final double codingCost()

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

Overrides:

codingCost in class ClassifierSplitModel

 public final double gainRatio()

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

 public final double infoGain()

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

 public final String leftSide(Instances data)

Prints left side of condition..

Parameters:

data - training set.

Overrides:

leftSide in class ClassifierSplitModel

 public final String rightSide(int index,
                               Instances data)

Prints the condition satisfied by instances in a subset.

Parameters:

index - of subset

data - training set.

Overrides:

rightSide in class ClassifierSplitModel

 public final 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".

Parameters:

index - index of the nominal value tested

data - the data containing instance structure info

Returns:

a value of type 'String'

Overrides:

sourceExpression in class ClassifierSplitModel

 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).

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

Returns the minsAndMaxs of the index.th subset.

 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.

Overrides:

weights in class ClassifierSplitModel

 public final int whichSubset(Instance instance) throws Exception

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

Throws: Exception

if something goes wrong

Overrides:

whichSubset in class ClassifierSplitModel

All Packages  Class Hierarchy  This Package  Previous  Next  Index  WEKA's home