weka.classifiers.trees.j48.C45PruneableClassifierTree Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of weka-dev Show documentation

The Waikato Environment for Knowledge Analysis (WEKA), a machine learning workbench. This version represents the developer version, the "bleeding edge" of development, you could say. New functionality gets added to this version.

There is a newer version: 3.9.6

Show newest version

/*
 *   This program is free software: you can redistribute it and/or modify
 *   it under the terms of the GNU General Public License as published by
 *   the Free Software Foundation, either version 3 of the License, or
 *   (at your option) any later version.
 *
 *   This program is distributed in the hope that it will be useful,
 *   but WITHOUT ANY WARRANTY; without even the implied warranty of
 *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *   GNU General Public License for more details.
 *
 *   You should have received a copy of the GNU General Public License
 *   along with this program.  If not, see .
 */

/*
 *    C45PruneableClassifierTree.java
 *    Copyright (C) 1999-2012 University of Waikato, Hamilton, New Zealand
 *
 */

package weka.classifiers.trees.j48;

import weka.core.Capabilities;
import weka.core.Capabilities.Capability;
import weka.core.Instances;
import weka.core.RevisionUtils;
import weka.core.Utils;

/**
 * Class for handling a tree structure that can
 * be pruned using C4.5 procedures.
 *
 * @author Eibe Frank ([email protected])
 * @version $Revision: 15122 $
 */

public class C45PruneableClassifierTree 
  extends ClassifierTree {

  /** for serialization */
  static final long serialVersionUID = -4813820170260388194L;
  
  /** True if the tree is to be pruned. */
  protected boolean m_pruneTheTree = false;
  
  /** True if the tree is to be collapsed. */
  protected boolean m_collapseTheTree = false;

  /** The confidence factor for pruning. */
  protected float m_CF = 0.25f;

  /** Is subtree raising to be performed? */
  protected boolean m_subtreeRaising = true;

  /** Cleanup after the tree has been built. */
  protected boolean m_cleanup = true;

  /**
   * Constructor for pruneable tree structure. Stores reference
   * to associated training data at each node.
   *
   * @param toSelectLocModel selection method for local splitting model
   * @param pruneTree true if the tree is to be pruned
   * @param cf the confidence factor for pruning
   * @param raiseTree
   * @param cleanup
   * @throws Exception if something goes wrong
   */
  public C45PruneableClassifierTree(ModelSelection toSelectLocModel,
				    boolean pruneTree,float cf,
				    boolean raiseTree,
				    boolean cleanup,
                                    boolean collapseTree)
       throws Exception {

    super(toSelectLocModel);

    m_pruneTheTree = pruneTree;
    m_CF = cf;
    m_subtreeRaising = raiseTree;
    m_cleanup = cleanup;
    m_collapseTheTree = collapseTree;
  }

  /**
   * Method for building a pruneable classifier tree.
   *
   * @param data the data for building the tree
   * @throws Exception if something goes wrong
   */
  public void buildClassifier(Instances data) throws Exception {

    // remove instances with missing class
    data = new Instances(data);
    data.deleteWithMissingClass();
    
   buildTree(data, m_subtreeRaising || !m_cleanup);
   if (m_collapseTheTree) {
     collapse();
   }
   if (m_pruneTheTree) {
     prune();
   }
   if (m_cleanup) {
     cleanup(new Instances(data, 0));
   }
  }

  /**
   * Collapses a tree to a node if training error doesn't increase.
   */
  public final void collapse(){

    double errorsOfSubtree;
    double errorsOfTree;
    int i;

    if (!m_isLeaf){
      errorsOfSubtree = getTrainingErrors();
      errorsOfTree = localModel().distribution().numIncorrect();
      if (errorsOfSubtree >= errorsOfTree-1E-3){

	// Free adjacent trees
	m_sons = null;
	m_isLeaf = true;
			
	// Get NoSplit Model for tree.
	m_localModel = new NoSplit(localModel().distribution());
      }else
	for (i=0;i