Package weka.classifiers.trees.ht

Class InfoGainSplitMetric

java.lang.Object
weka.classifiers.trees.ht.SplitMetric
weka.classifiers.trees.ht.InfoGainSplitMetric
All Implemented Interfaces:
Serializable
public class InfoGainSplitMetric extends SplitMetric implements Serializable
Implements the info gain splitting criterion
Version:
$Revision: 9720 $
Author:
Richard Kirkby (rkirkby@cs.waikato.ac.nz), Mark Hall (mhall{[at]}pentaho{[dot]}com)
See Also:

  • Constructor Details

    • InfoGainSplitMetric

      public InfoGainSplitMetric(double minFracWeightForTwoBranches)

  • Method Details

    • evaluateSplit

      public double evaluateSplit(Map<String,WeightMass> preDist, List<Map<String,WeightMass>> postDist)
      Description copied from class: SplitMetric
      Evaluate the merit of a split
      Specified by:
      evaluateSplit in class SplitMetric
      Parameters:
      preDist - the class distribution before the split
      postDist - the class distributions after the split
      Returns:
      the merit of the split

    • getMetricRange

      public double getMetricRange(Map<String,WeightMass> preDist)
      Description copied from class: SplitMetric
      Get the range of the splitting metric
      Specified by:
      getMetricRange in class SplitMetric
      Parameters:
      preDist - the pre-split class distribution
      Returns:
      the range of the splitting metric