weka.classifiers.trees
Class DecisionStump

java.lang.Object
  weka.classifiers.Classifier
      weka.classifiers.trees.DecisionStump

All Implemented Interfaces:

java.lang.Cloneable, OptionHandler, java.io.Serializable, Sourcable, WeightedInstancesHandler

public class DecisionStump

extends Classifier

implements WeightedInstancesHandler, Sourcable

Class for building and using a decision stump. Usually used in conjunction with a boosting algorithm. Typical usage:

java weka.classifiers.trees.LogitBoost -I 100 -W weka.classifiers.trees.DecisionStump -t training_data

Version:

$Revision: 1.18 $

Author:

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

See Also:

Serialized Form

Field Summary

private int	m_AttIndex The attribute used for classification.
private double[][]	m_Distribution The distribution of class values or the means in each subset.
private Instances	m_Instances The instances used for training.
private double	m_SplitPoint The split point (index respectively).

Fields inherited from class weka.classifiers.Classifier

m_Debug

Constructor Summary

DecisionStump()

Method Summary

void	buildClassifier(Instances instances) Generates the classifier.
double[]	distributionForInstance(Instance instance) Calculates the class membership probabilities for the given test instance.
private double	findSplitNominal(int index) Finds best split for nominal attribute and returns value.
private double	findSplitNominalNominal(int index) Finds best split for nominal attribute and nominal class and returns value.
private double	findSplitNominalNumeric(int index) Finds best split for nominal attribute and numeric class and returns value.
private double	findSplitNumeric(int index) Finds best split for numeric attribute and returns value.
private double	findSplitNumericNominal(int index) Finds best split for numeric attribute and nominal class and returns value.
private double	findSplitNumericNumeric(int index) Finds best split for numeric attribute and numeric class and returns value.
java.lang.String	globalInfo() Returns a string describing classifier
static void	main(java.lang.String[] argv) Main method for testing this class.
private java.lang.String	printClass(double[] dist) Prints a classification.
private java.lang.String	printDist(double[] dist) Prints a class distribution.
private java.lang.String	sourceClass(Attribute c, double[] dist)
java.lang.String	toSource(java.lang.String className) Returns the decision tree as Java source code.
java.lang.String	toString() Returns a description of the classifier.
private double	variance(double[][] s, double[] sS, double[] sumOfWeights) Computes variance for subsets.
private int	whichSubset(Instance instance) Returns the subset an instance falls into.

Methods inherited from class weka.classifiers.Classifier

classifyInstance, debugTipText, forName, getDebug, getOptions, listOptions, makeCopies, setDebug, setOptions

Methods inherited from class java.lang.Object

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

Field Detail

m_AttIndex

private int m_AttIndex

The attribute used for classification.

m_SplitPoint

private double m_SplitPoint

The split point (index respectively).

m_Distribution

private double[][] m_Distribution

The distribution of class values or the means in each subset.

m_Instances

private Instances m_Instances

The instances used for training.

Constructor Detail

DecisionStump

public DecisionStump()

Method Detail

globalInfo

public java.lang.String globalInfo()

Returns a string describing classifier

Returns:

a description suitable for displaying in the explorer/experimenter gui

buildClassifier

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

Generates the classifier.

Specified by:

buildClassifier in class Classifier

Parameters:

instances - set of instances serving as training data

Throws:

java.lang.Exception - if the classifier has not been generated successfully

distributionForInstance

public double[] distributionForInstance(Instance instance)
                                 throws java.lang.Exception

Calculates the class membership probabilities for the given test instance.

Overrides:

distributionForInstance in class Classifier

Parameters:

instance - the instance to be classified

Returns:

predicted class probability distribution

Throws:

java.lang.Exception - if distribution can't be computed

toSource

public java.lang.String toSource(java.lang.String className)
                          throws java.lang.Exception

Returns the decision tree as Java source code.

Specified by:

toSource in interface Sourcable

Parameters:

className - the name that should be given to the source class.

Returns:

the tree as Java source code

Throws:

java.lang.Exception - if something goes wrong

sourceClass

private java.lang.String sourceClass(Attribute c,
                                     double[] dist)

toString

public java.lang.String toString()

Returns a description of the classifier.

Returns:

a description of the classifier as a string.

printDist

private java.lang.String printDist(double[] dist)
                            throws java.lang.Exception

Prints a class distribution.

Parameters:

dist - the class distribution to print

Returns:

the distribution as a string

Throws:

java.lang.Exception - if distribution can't be printed

printClass

private java.lang.String printClass(double[] dist)
                             throws java.lang.Exception

Prints a classification.

Parameters:

dist - the class distribution

Returns:

the classificationn as a string

Throws:

java.lang.Exception - if the classification can't be printed

findSplitNominal

private double findSplitNominal(int index)
                         throws java.lang.Exception

Finds best split for nominal attribute and returns value.

Parameters:

index - attribute index

Returns:

value of criterion for the best split

Throws:

java.lang.Exception - if something goes wrong

findSplitNominalNominal

private double findSplitNominalNominal(int index)
                                throws java.lang.Exception

Finds best split for nominal attribute and nominal class and returns value.

Parameters:

index - attribute index

Returns:

value of criterion for the best split

Throws:

java.lang.Exception - if something goes wrong

findSplitNominalNumeric

private double findSplitNominalNumeric(int index)
                                throws java.lang.Exception

Finds best split for nominal attribute and numeric class and returns value.

Parameters:

index - attribute index

Returns:

value of criterion for the best split

Throws:

java.lang.Exception - if something goes wrong

findSplitNumeric

private double findSplitNumeric(int index)
                         throws java.lang.Exception

Finds best split for numeric attribute and returns value.

Parameters:

index - attribute index

Returns:

value of criterion for the best split

Throws:

java.lang.Exception - if something goes wrong

findSplitNumericNominal

private double findSplitNumericNominal(int index)
                                throws java.lang.Exception

Finds best split for numeric attribute and nominal class and returns value.

Parameters:

index - attribute index

Returns:

value of criterion for the best split

Throws:

java.lang.Exception - if something goes wrong

findSplitNumericNumeric

private double findSplitNumericNumeric(int index)
                                throws java.lang.Exception

Finds best split for numeric attribute and numeric class and returns value.

Parameters:

index - attribute index

Returns:

value of criterion for the best split

Throws:

java.lang.Exception - if something goes wrong

variance

private double variance(double[][] s,
                        double[] sS,
                        double[] sumOfWeights)

Computes variance for subsets.

whichSubset

private int whichSubset(Instance instance)
                 throws java.lang.Exception

Returns the subset an instance falls into.

Throws:

java.lang.Exception

main

public static void main(java.lang.String[] argv)

Main method for testing this class.

Parameters:

argv - the options