m5base.java

代码是一个分类器的实现,其中使用了部分weka的源代码。可以将项目导入eclipse运行

/*
 *    M5Base.java
 *    Copyright (C) 2000 Mark Hall
 *
 *    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 2 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, write to the Free Software
 *    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 */
package weka.classifiers.trees.m5;

import weka.classifiers.Classifier;
import weka.classifiers.functions.LinearRegression;
import weka.core.AdditionalMeasureProducer;
import weka.core.Capabilities;
import weka.core.FastVector;
import weka.core.Instance;
import weka.core.Instances;
import weka.core.Option;
import weka.core.OptionHandler;
import weka.core.TechnicalInformation;
import weka.core.TechnicalInformationHandler;
import weka.core.Utils;
import weka.core.TechnicalInformation.Field;
import weka.core.TechnicalInformation.Type;
import weka.filters.Filter;
import weka.filters.supervised.attribute.NominalToBinary;
import weka.filters.unsupervised.attribute.RemoveUseless;
import weka.filters.unsupervised.attribute.ReplaceMissingValues;

import java.util.Enumeration;
import java.util.Random;
import java.util.Vector;

/**
 * M5Base. Implements base routines
 * for generating M5 Model trees and rules. <p>
 * 
 * The original algorithm M5 was invented by Quinlan: <br/>
 * 
 * Quinlan J. R. (1992). Learning with continuous classes. Proceedings of
 * the Australian Joint Conference on Artificial Intelligence. 343--348.
 * World Scientific, Singapore. <p/>
 * 
 * Yong Wang made improvements and created M5': <br/>
 * 
 * Wang, Y and Witten, I. H. (1997). Induction of model trees for
 * predicting continuous classes. Proceedings of the poster papers of the
 * European Conference on Machine Learning. University of Economics,
 * Faculty of Informatics and Statistics, Prague. <p/>
 *
 * Valid options are:<p>
 * 
 * -U <br>
 * Use unsmoothed predictions. <p>
 *
 * -R <br>
 * Build regression tree/rule rather than model tree/rule
 *
 * @version $Revision: 1.16 $
 */
public abstract class M5Base 
  extends Classifier 
  implements OptionHandler,
	     AdditionalMeasureProducer,
	     TechnicalInformationHandler {

  /**
   * the instances covered by the tree/rules
   */
  private Instances		     m_instances;

  /**
   * the rule set
   */
  protected FastVector		     m_ruleSet;

  /**
   * generate a decision list instead of a single tree.
   */
  private boolean		     m_generateRules;

  /**
   * use unsmoothed predictions
   */
  private boolean		     m_unsmoothedPredictions;

  /**
   * filter to fill in missing values
   */
  private ReplaceMissingValues m_replaceMissing;

  /**
   * filter to convert nominal attributes to binary
   */
  private NominalToBinary      m_nominalToBinary;
  
  /**
   * for removing useless attributes 
   */
  private RemoveUseless m_removeUseless;

  /**
   * Save instances at each node in an M5 tree for visualization purposes.
   */
  protected boolean m_saveInstances = false;

  /**
   * Make a regression tree/rule instead of a model tree/rule
   */
  protected boolean m_regressionTree;

  /**
   * Do not prune tree/rules
   */
  protected boolean m_useUnpruned = false;

  /**
   * The minimum number of instances to allow at a leaf node
   */
  protected double m_minNumInstances = 4;

  /**
   * Constructor
   */
  public M5Base() {
    m_generateRules = false;
    m_unsmoothedPredictions = false;
    m_useUnpruned = false;
    m_minNumInstances = 4;
  }

  /**
   * returns information about the classifier
   * @return a description suitable for
   * displaying in the explorer/experimenter gui
   */
  public String globalInfo() {
    return 
        "M5Base. Implements base routines for generating M5 Model trees and " 
      + "rules\n"
      + "The original algorithm M5 was invented by R. Quinlan and Yong Wang "
      + "made improvements.\n\n"
      + "For more information see:\n\n"
      + getTechnicalInformation().toString();
  }

  /**
   * Returns an instance of a TechnicalInformation object, containing 
   * detailed information about the technical background of this class,
   * e.g., paper reference or book this class is based on.
   * 
   * @return the technical information about this class
   */
  public TechnicalInformation getTechnicalInformation() {
    TechnicalInformation 	result;
    TechnicalInformation 	additional;
    
    result = new TechnicalInformation(Type.INPROCEEDINGS);
    result.setValue(Field.AUTHOR, "Ross J. Quinlan");
    result.setValue(Field.TITLE, "Learning with Continuous Classes");
    result.setValue(Field.BOOKTITLE, "5th Australian Joint Conference on Artificial Intelligence");
    result.setValue(Field.YEAR, "1992");
    result.setValue(Field.PAGES, "343-348");
    result.setValue(Field.PUBLISHER, "World Scientific");
    result.setValue(Field.ADDRESS, "Singapore");
    
    additional = result.add(Type.INPROCEEDINGS);
    additional.setValue(Field.AUTHOR, "Y. Wang and I. H. Witten");
    additional.setValue(Field.TITLE, "Induction of model trees for predicting continuous classes");
    additional.setValue(Field.BOOKTITLE, "Poster papers of the 9th European Conference on Machine Learning");
    additional.setValue(Field.YEAR, "1997");
    additional.setValue(Field.PUBLISHER, "Springer");
    
    return result;
  }

  /**
   * Returns an enumeration describing the available options
   * 
   * @return an enumeration of all the available options
   */
  public Enumeration listOptions() {
    Vector newVector = new Vector(4);

    newVector.addElement(new Option("\tUse unpruned tree/rules", 
				    "N", 0, "-N"));

    newVector.addElement(new Option("\tUse unsmoothed predictions", 
				    "U", 0, "-U"));

    newVector.addElement(new Option("\tBuild regression tree/rule rather "
				    +"than a model tree/rule", 
				    "R", 0, "-R"));

    newVector.addElement(new Option("\tSet minimum number of instances "
				    +"per leaf\n\t(default 4)",
				    "M",1,"-M <minimum number of instances>"));
    return newVector.elements();
  } 

  /**
   * Parses a given list of options. <p/>
   * 
   * Valid options are:<p>
   * 
   * -U <br>
   * Use unsmoothed predictions. <p>
   *
   * -R <br>
   * Build a regression tree rather than a model tree. <p>
   * 
   * @param options the list of options as an array of strings
   * @throws Exception if an option is not supported
   */
  public void setOptions(String[] options) throws Exception {
    setUnpruned(Utils.getFlag('N', options));
    setUseUnsmoothed(Utils.getFlag('U', options));
    setBuildRegressionTree(Utils.getFlag('R', options));
    String optionString = Utils.getOption('M', options);
    if (optionString.length() != 0) {
      setMinNumInstances((new Double(optionString)).doubleValue());
    }
    Utils.checkForRemainingOptions(options);
  } 

  /**
   * Gets the current settings of the classifier.
   * 
   * @return an array of strings suitable for passing to setOptions
   */
  public String[] getOptions() {
    String[] options = new String[5];
    int current = 0;

    if (getUnpruned()) {
      options[current++] = "-N";
    }

    if (getUseUnsmoothed()) {
      options[current++] = "-U";
    } 

    if (getBuildRegressionTree()) {
      options[current++] = "-R";
    }

    options[current++] = "-M"; 
    options[current++] = ""+getMinNumInstances();

    while (current < options.length) {
      options[current++] = "";
    } 
    return options;
  } 

  /**
   * Returns the tip text for this property
   * 
   * @return 		tip text for this property suitable for
   * 			displaying in the explorer/experimenter gui
   */
  public String unprunedTipText() {
    return "Whether unpruned tree/rules are to be generated.";
  }

  /**
   * Use unpruned tree/rules
   *
   * @param unpruned true if unpruned tree/rules are to be generated
   */
  public void setUnpruned(boolean unpruned) {
    m_useUnpruned = unpruned;
  }

  /**
   * Get whether unpruned tree/rules are being generated
   *
   * @return true if unpruned tree/rules are to be generated
   */
  public boolean getUnpruned() {
    return m_useUnpruned;
  }

  /**
   * Returns the tip text for this property
   * 
   * @return 		tip text for this property suitable for
   * 			displaying in the explorer/experimenter gui
   */
  public String generateRulesTipText() {
    return "Whether to generate rules (decision list) rather than a tree.";
  }

  /**
   * Generate rules (decision list) rather than a tree
   * 
   * @param u true if rules are to be generated
   */
  protected void setGenerateRules(boolean u) {
    m_generateRules = u;
  } 

  /**
   * get whether rules are being generated rather than a tree
   * 
   * @return true if rules are to be generated
   */