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

Class weka.classifiers.j48.InfoGainSplitCrit

java.lang.Object
   |
   +----weka.classifiers.j48.SplitCriterion
           |
           +----weka.classifiers.j48.EntropyBasedSplitCrit
                   |
                   +----weka.classifiers.j48.InfoGainSplitCrit

public final class InfoGainSplitCrit
extends EntropyBasedSplitCrit
Class for computing the information gain for a given distribution.

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

Constructor Index

 o InfoGainSplitCrit()

Method Index

 o splitCritValue(Distribution)
This method is a straightforward implementation of the information gain criterion for the given distribution.
 o splitCritValue(Distribution, double)
This method computes the information gain in the same way C4.5 does.
 o splitCritValue(Distribution, double, double)
This method computes the information gain in the same way C4.5 does.

Constructors

 o InfoGainSplitCrit
 public InfoGainSplitCrit()

Methods

 o splitCritValue
 public final double splitCritValue(Distribution bags)
This method is a straightforward implementation of the information gain criterion for the given distribution.

Overrides:
splitCritValue in class SplitCriterion
 o splitCritValue
 public final double splitCritValue(Distribution bags,
                                    double totalNoInst)
This method computes the information gain in the same way C4.5 does.

Parameters:
distribution - the distribution
totalNoInst - weight of ALL instances (including the ones with missing values).
 o splitCritValue
 public final double splitCritValue(Distribution bags,
                                    double totalNoInst,
                                    double oldEnt)
This method computes the information gain in the same way C4.5 does.

Parameters:
distribution - the distribution
totalNoInst - weight of ALL instances
oldEnt - entropy with respect to "no-split"-model.

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