weka.classifiers.trees.j48
Class SplitCriterion

java.lang.Object
  weka.classifiers.trees.j48.SplitCriterion

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

EntropyBasedSplitCrit

public abstract class SplitCriterion

extends java.lang.Object

implements java.io.Serializable

Abstract class for computing splitting criteria with respect to distributions of class values.

Version:

$Revision: 1.5 $

Author:

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

See Also:

Serialized Form

Constructor Summary

SplitCriterion()

Method Summary

double	splitCritValue(Distribution bags) Computes result of splitting criterion for given distribution.
double	splitCritValue(Distribution train, Distribution test) Computes result of splitting criterion for given training and test distributions.
double	splitCritValue(Distribution train, Distribution test, Distribution defC) Computes result of splitting criterion for given training and test distributions and given default distribution.
double	splitCritValue(Distribution train, Distribution test, int noClassesDefault) Computes result of splitting criterion for given training and test distributions and given number of classes.

Methods inherited from class java.lang.Object

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

Constructor Detail

SplitCriterion

public SplitCriterion()

Method Detail

splitCritValue

public double splitCritValue(Distribution bags)

Computes result of splitting criterion for given distribution.

Returns:

value of splitting criterion. 0 by default

splitCritValue

public double splitCritValue(Distribution train,
                             Distribution test)

Computes result of splitting criterion for given training and test distributions.

Returns:

value of splitting criterion. 0 by default

splitCritValue

public double splitCritValue(Distribution train,
                             Distribution test,
                             int noClassesDefault)

Computes result of splitting criterion for given training and test distributions and given number of classes.

Returns:

value of splitting criterion. 0 by default

splitCritValue

public double splitCritValue(Distribution train,
                             Distribution test,
                             Distribution defC)

Computes result of splitting criterion for given training and test distributions and given default distribution.

Returns:

value of splitting criterion. 0 by default