• Home
• nz.ac.waikato.cms.weka
• weka-stable
• 3.8.0
• source code
• KStar.java

×

Project download

Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance.
Project price only 1 $

You can buy this project and download/modify it how often you want.

weka.classifiers.lazy.KStar Maven / Gradle / Ivy

/*
 *   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 .
 */

/*
 *    KStar.java
 *    Copyright (C) 1995-2012 University of Waikato, Hamilton, New Zealand
 *
 */

package weka.classifiers.lazy;

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

import weka.classifiers.AbstractClassifier;
import weka.classifiers.UpdateableClassifier;
import weka.classifiers.lazy.kstar.KStarCache;
import weka.classifiers.lazy.kstar.KStarConstants;
import weka.classifiers.lazy.kstar.KStarNominalAttribute;
import weka.classifiers.lazy.kstar.KStarNumericAttribute;
import weka.core.Attribute;
import weka.core.Capabilities;
import weka.core.Capabilities.Capability;
import weka.core.Instance;
import weka.core.Instances;
import weka.core.Option;
import weka.core.RevisionUtils;
import weka.core.SelectedTag;
import weka.core.Tag;
import weka.core.TechnicalInformation;
import weka.core.TechnicalInformation.Field;
import weka.core.TechnicalInformation.Type;
import weka.core.TechnicalInformationHandler;
import weka.core.Utils;

/**
 
 * K* is an instance-based classifier, that is the class of a test instance is based upon the class of those training instances similar to it, as determined by some similarity function.  It differs from other instance-based learners in that it uses an entropy-based distance function.

 * 

 * For more information on K*, see

 * 

 * John G. Cleary, Leonard E. Trigg: K*: An Instance-based Learner Using an Entropic Distance Measure. In: 12th International Conference on Machine Learning, 108-114, 1995.
 * 

 
 *
 
 * BibTeX:
 * 
 * @inproceedings{Cleary1995,
 *    author = {John G. Cleary and Leonard E. Trigg},
 *    booktitle = {12th International Conference on Machine Learning},
 *    pages = {108-114},
 *    title = {K*: An Instance-based Learner Using an Entropic Distance Measure},
 *    year = {1995}
 * }
 * 
 * 

 
 *
 
 * Valid options are: 

 * 
 * 
 -B <num>
 *  Manual blend setting (default 20%)
 * 
 * 
 * 
 -E
 *  Enable entropic auto-blend setting (symbolic class only)
 * 
 * 
 * 
 -M <char>
 *  Specify the missing value treatment mode (default a)
 *  Valid options are: a(verage), d(elete), m(axdiff), n(ormal)
 * 
 * 
 
 *
 * @author Len Trigg ([email protected])
 * @author Abdelaziz Mahoui ([email protected]) - Java port
 * @version $Revision: 10141 $
 */
public class KStar 
  extends AbstractClassifier
  implements KStarConstants, UpdateableClassifier, TechnicalInformationHandler {

  /** for serialization */
  static final long serialVersionUID = 332458330800479083L;
  
  /** The training instances used for classification. */
  protected Instances m_Train; 

  /** The number of instances in the dataset */
  protected int m_NumInstances;

  /** The number of class values */
  protected int m_NumClasses;

  /** The number of attributes */
  protected int m_NumAttributes;

  /** The class attribute type */
  protected int m_ClassType;

  /** Table of random class value colomns */
  protected int [][] m_RandClassCols;

  /** Flag turning on and off the computation of random class colomns */
  protected int m_ComputeRandomCols = ON;

  /** Flag turning on and off the initialisation of config variables */
  protected int m_InitFlag = ON;

  /**
   * A custom data structure for caching distinct attribute values
   * and their scale factor or stop parameter.
   */
  protected KStarCache [] m_Cache;

  /** missing value treatment */
  protected int m_MissingMode = M_AVERAGE;

  /** 0 = use specified blend, 1 = entropic blend setting */
  protected int m_BlendMethod = B_SPHERE;

  /** default sphere of influence blend setting */
  protected int m_GlobalBlend = 20;

  /** Define possible missing value handling methods */
  public static final Tag [] TAGS_MISSING = {
    new Tag(M_DELETE, "Ignore the instances with missing values"),
    new Tag(M_MAXDIFF, "Treat missing values as maximally different"),
    new Tag(M_NORMAL, "Normalize over the attributes"),
    new Tag(M_AVERAGE, "Average column entropy curves")
      };
    
  /**
   * Returns a string describing classifier
   * @return a description suitable for
   * displaying in the explorer/experimenter gui
   */
  public String globalInfo() {

    return "K* is an instance-based classifier, that is the class of a test "
      + "instance is based upon the class of those training instances "
      + "similar to it, as determined by some similarity function.  It differs "
      + "from other instance-based learners in that it uses an entropy-based "
      + "distance function.\n\n"
      + "For more information on K*, 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;
    
    result = new TechnicalInformation(Type.INPROCEEDINGS);
    result.setValue(Field.AUTHOR, "John G. Cleary and Leonard E. Trigg");
    result.setValue(Field.TITLE, "K*: An Instance-based Learner Using an Entropic Distance Measure");
    result.setValue(Field.BOOKTITLE, "12th International Conference on Machine Learning");
    result.setValue(Field.YEAR, "1995");
    result.setValue(Field.PAGES, "108-114");
    
    return result;
  }

  /**
   * Returns default capabilities of the classifier.
   *
   * @return      the capabilities of this classifier
   */
  public Capabilities getCapabilities() {
    Capabilities result = super.getCapabilities();
    result.disableAll();

    // attributes
    result.enable(Capability.NOMINAL_ATTRIBUTES);
    result.enable(Capability.NUMERIC_ATTRIBUTES);
    result.enable(Capability.DATE_ATTRIBUTES);
    result.enable(Capability.MISSING_VALUES);

    // class
    result.enable(Capability.NOMINAL_CLASS);
    result.enable(Capability.NUMERIC_CLASS);
    result.enable(Capability.DATE_CLASS);
    result.enable(Capability.MISSING_CLASS_VALUES);

    // instances
    result.setMinimumNumberInstances(0);
    
    return result;
  }

  /**
   * Generates the classifier.
   *
   * @param instances set of instances serving as training data 
   * @throws Exception if the classifier has not been generated successfully
   */
  public void buildClassifier(Instances instances) throws Exception {
    
    // can classifier handle the data?
    getCapabilities().testWithFail(instances);

    // remove instances with missing class
    instances = new Instances(instances);
    instances.deleteWithMissingClass();
    
    m_Train = new Instances(instances, 0, instances.numInstances());

    // initializes class attributes ** java-speaking! :-) **
    init_m_Attributes();
  }
  
  /**
   * Adds the supplied instance to the training set
   *
   * @param instance the instance to add
   * @throws Exception if instance could not be incorporated successfully
   */
  public void updateClassifier(Instance instance) throws Exception {
    
    if (m_Train.equalHeaders(instance.dataset()) == false)
      throw new Exception("Incompatible instance types\n" + m_Train.equalHeadersMsg(instance.dataset()));
    if ( instance.classIsMissing() )
      return;
    m_Train.add(instance);
    // update relevant attributes ...
    update_m_Attributes();
  }

  /**
   * Calculates the class membership probabilities for the given test instance.
   *
   * @param instance the instance to be classified
   * @return predicted class probability distribution
   * @throws Exception if an error occurred during the prediction
   */
  public double [] distributionForInstance(Instance instance) throws Exception {

    double transProb = 0.0, temp = 0.0;
    double [] classProbability = new double[m_NumClasses];
    double [] predictedValue = new double[1];

    // initialization ...
    for (int i=0; i enu = m_Train.enumerateInstances();
    while ( enu.hasMoreElements() ) {
      trainInstance = (Instance)enu.nextElement();
      transProb = instanceTransformationProbability(instance, trainInstance);      
      switch ( m_ClassType )
	{
	case Attribute.NOMINAL:
	  classProbability[(int)trainInstance.classValue()] += transProb;
	  break;
	case Attribute.NUMERIC:
	  predictedValue[0] += transProb * trainInstance.classValue();
	  temp += transProb;
	  break;
	}
    }
    if (m_ClassType == Attribute.NOMINAL) {
      double sum = Utils.sum(classProbability);
      if (sum <= 0.0)
	for (int i=0; i listOptions() {

    Vector optVector = new Vector( 3 );
    optVector.addElement(new Option(
	      "\tManual blend setting (default 20%)\n",
	      "B", 1, "-B "));
    optVector.addElement(new Option(
	      "\tEnable entropic auto-blend setting (symbolic class only)\n",
	      "E", 0, "-E"));
    optVector.addElement(new Option(
	      "\tSpecify the missing value treatment mode (default a)\n"
	      +"\tValid options are: a(verage), d(elete), m(axdiff), n(ormal)\n",
	      "M", 1,"-M "));
    
    optVector.addAll(Collections.list(super.listOptions()));
    
    return optVector.elements();
  }
   
  /**
   * Returns the tip text for this property
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String globalBlendTipText() {
    return "The parameter for global blending. Values are restricted to [0,100].";
  }

  /**
   * Set the global blend parameter
   * @param b the value for global blending
   */
  public void setGlobalBlend(int b) {
     m_GlobalBlend = b;
      if ( m_GlobalBlend > 100 ) {
	m_GlobalBlend = 100;
      }
      if ( m_GlobalBlend < 0 ) {
	m_GlobalBlend = 0;
      }
  }

  /**
   * Get the value of the global blend parameter
   * @return the value of the global blend parameter
   */
  public int getGlobalBlend() {
    return m_GlobalBlend;
  }
   
  /**
   * Returns the tip text for this property
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String entropicAutoBlendTipText() {
    return "Whether entropy-based blending is to be used.";
  }

  /**
   * Set whether entropic blending is to be used.
   * @param e true if entropic blending is to be used
   */
  public void setEntropicAutoBlend(boolean e) {
    if (e) {
      m_BlendMethod = B_ENTROPY;
    } else {
      m_BlendMethod = B_SPHERE;
    }
  }

  /**
   * Get whether entropic blending being used
   * @return true if entropic blending is used
   */
  public boolean getEntropicAutoBlend() {
    if (m_BlendMethod == B_ENTROPY) {
      return true;
    }

    return false;
  }

  /**
   * Parses a given list of options. 

   *
   
   * Valid options are: 

   * 
   * 
 -B <num>
   *  Manual blend setting (default 20%)
   * 
   * 
   * 
 -E
   *  Enable entropic auto-blend setting (symbolic class only)
   * 
   * 
   * 
 -M <char>
   *  Specify the missing value treatment mode (default a)
   *  Valid options are: a(verage), d(elete), m(axdiff), n(ormal)
   * 
   * 
   
   *
   * @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 {
    
    String blendStr = Utils.getOption('B', options);
    if (blendStr.length() != 0) {
      setGlobalBlend(Integer.parseInt(blendStr));
    }

    setEntropicAutoBlend(Utils.getFlag('E', options));
    
    String missingModeStr = Utils.getOption('M', options);
    if (missingModeStr.length() != 0) {
      switch ( missingModeStr.charAt(0) ) {
      case 'a':
	setMissingMode(new SelectedTag(M_AVERAGE, TAGS_MISSING));
	break;
      case 'd':
	setMissingMode(new SelectedTag(M_DELETE, TAGS_MISSING));
	break;
      case 'm':
	setMissingMode(new SelectedTag(M_MAXDIFF, TAGS_MISSING));
	break;
      case 'n':
	setMissingMode(new SelectedTag(M_NORMAL, TAGS_MISSING));
	break;
      default:
	setMissingMode(new SelectedTag(M_AVERAGE, TAGS_MISSING));
      }
    }
    
    super.setOptions(options);
    
    Utils.checkForRemainingOptions(options);
  }


  /**
   * Gets the current settings of K*.
   *
   * @return an array of strings suitable for passing to setOptions()
   */
  public String [] getOptions() {
    // -B  -E -M 
    Vector options = new Vector();
    
    options.add("-B");
    options.add("" + m_GlobalBlend);

    if (getEntropicAutoBlend()) {
        options.add("-E");
    }

    options.add("-M");
    if (m_MissingMode == M_AVERAGE) {
        options.add("" + "a");
    }
    else if (m_MissingMode == M_DELETE) {
        options.add("" + "d");
    }
    else if (m_MissingMode == M_MAXDIFF) {
        options.add("" + "m");
    }
    else if (m_MissingMode == M_NORMAL) {
        options.add("" + "n");
    }
    
    Collections.addAll(options, super.getOptions());
    
    return options.toArray(new String[0]);
  }

  /**
   * Returns a description of this classifier.
   *
   * @return a description of this classifier as a string.
   */
  public String toString() {
    StringBuffer st = new StringBuffer();
    st.append("KStar Beta Verion (0.1b).\n"
	      +"Copyright (c) 1995-97 by Len Trigg ().\n"
	      +"Java port to Weka by Abdelaziz Mahoui "
	      +"().\n\nKStar options : ");
    String [] ops = getOptions();
    for (int i=0;i 0; j--) {
      index = (int) ( generator.nextDouble() * (double)j );
      temp = newArray[j];
      newArray[j] = newArray[index];
      newArray[index] = temp;
    }
    return newArray;
  }
  
  /**
   * Returns the revision string.
   * 
   * @return		the revision
   */
  public String getRevision() {
    return RevisionUtils.extract("$Revision: 10141 $");
  }

} // class end