mioptimalball.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

/*
 *    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.
 */

/*
 * MIOptimalBall.java
 * Copyright (C) 2005 University of Waikato, Hamilton, New Zealand
 *
 */

package weka.classifiers.mi;

import weka.classifiers.Classifier;
import weka.core.Capabilities;
import weka.core.Instance;
import weka.core.Instances;
import weka.core.MultiInstanceCapabilitiesHandler;
import weka.core.Option;
import weka.core.OptionHandler;
import weka.core.SelectedTag;
import weka.core.Tag;
import weka.core.TechnicalInformation;
import weka.core.TechnicalInformationHandler;
import weka.core.Utils;
import weka.core.WeightedInstancesHandler;
import weka.core.Capabilities.Capability;
import weka.core.TechnicalInformation.Field;
import weka.core.TechnicalInformation.Type;
import weka.core.matrix.DoubleVector;
import weka.filters.Filter;
import weka.filters.unsupervised.attribute.MultiInstanceToPropositional;
import weka.filters.unsupervised.attribute.Normalize;
import weka.filters.unsupervised.attribute.PropositionalToMultiInstance;
import weka.filters.unsupervised.attribute.Standardize;

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

/**
 <!-- globalinfo-start -->
 * This classifier tries to find a suitable ball in the multiple-instance space, with a certain data point in the instance space as a ball center. The possible ball center is a certain instance in a positive bag. The possible radiuses are those which can achieve the highest classification accuracy. The model selects the maximum radius as the radius of the optimal ball.<br/>
 * <br/>
 * For more information about this algorithm, see:<br/>
 * <br/>
 * Peter Auer, Ronald Ortner: A Boosting Approach to Multiple Instance Learning. In: 15th European Conference on Machine Learning, 63-74, 2004.
 * <p/>
 <!-- globalinfo-end -->
 *
 <!-- technical-bibtex-start -->
 * BibTeX:
 * <pre>
 * &#64;inproceedings{Auer2004,
 *    author = {Peter Auer and Ronald Ortner},
 *    booktitle = {15th European Conference on Machine Learning},
 *    note = {LNAI 3201},
 *    pages = {63-74},
 *    publisher = {Springer},
 *    title = {A Boosting Approach to Multiple Instance Learning},
 *    year = {2004}
 * }
 * </pre>
 * <p/>
 <!-- technical-bibtex-end -->
 * 
 <!-- options-start -->
 * Valid options are: <p/>
 * 
 * <pre> -N &lt;num&gt;
 *  Whether to 0=normalize/1=standardize/2=neither. 
 *  (default 0=normalize)</pre>
 * 
 <!-- options-end -->
 * 
 * @author Lin Dong ([e-mail])
 * @version $Revision: 1.4 $ 
 */
public class MIOptimalBall 
  extends Classifier 
  implements OptionHandler, WeightedInstancesHandler, 
             MultiInstanceCapabilitiesHandler, TechnicalInformationHandler {  

  /** for serialization */
  static final long serialVersionUID = -6465750129576777254L;
  
  /** center of the optimal ball */
  protected double[] m_Center;

  /** radius of the optimal ball */
  protected double m_Radius;

  /** the distances from each instance in a positive bag to each bag*/
  protected double [][][]m_Distance;

  /** The filter used to standardize/normalize all values. */
  protected Filter m_Filter = null;

  /** Whether to normalize/standardize/neither */
  protected int m_filterType = FILTER_NORMALIZE;

  /** Normalize training data */
  public static final int FILTER_NORMALIZE = 0;
  /** Standardize training data */
  public static final int FILTER_STANDARDIZE = 1;
  /** No normalization/standardization */
  public static final int FILTER_NONE = 2;
  /** The filter to apply to the training data */
  public static final Tag [] TAGS_FILTER = {
    new Tag(FILTER_NORMALIZE, "Normalize training data"),
    new Tag(FILTER_STANDARDIZE, "Standardize training data"),
    new Tag(FILTER_NONE, "No normalization/standardization"),
  };

  /** filter used to convert the MI dataset into single-instance dataset */
  protected MultiInstanceToPropositional m_ConvertToSI = new MultiInstanceToPropositional();

  /** filter used to convert the single-instance dataset into MI dataset */
  protected PropositionalToMultiInstance m_ConvertToMI = new PropositionalToMultiInstance();

  /**
   * Returns a string describing this filter
   *
   * @return a description of the filter suitable for
   * displaying in the explorer/experimenter gui
   */
  public String globalInfo() {
    return
         "This classifier tries to find a suitable ball in the "
       + "multiple-instance space, with a certain data point in the instance "
       + "space as a ball center. The possible ball center is a certain "
       + "instance in a positive bag. The possible radiuses are those which can "
       + "achieve the highest classification accuracy. The model selects the "
       + "maximum radius as the radius of the optimal ball.\n\n"
       + "For more information about this algorithm, 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, "Peter Auer and Ronald Ortner");
    result.setValue(Field.TITLE, "A Boosting Approach to Multiple Instance Learning");
    result.setValue(Field.BOOKTITLE, "15th European Conference on Machine Learning");
    result.setValue(Field.YEAR, "2004");
    result.setValue(Field.PAGES, "63-74");
    result.setValue(Field.PUBLISHER, "Springer");
    result.setValue(Field.NOTE, "LNAI 3201");
    
    return result;
  }

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

    // attributes
    result.enable(Capability.NOMINAL_ATTRIBUTES);
    result.enable(Capability.RELATIONAL_ATTRIBUTES);
    result.enable(Capability.MISSING_VALUES);

    // class
    result.enable(Capability.BINARY_CLASS);
    result.enable(Capability.MISSING_CLASS_VALUES);
    
    // other
    result.enable(Capability.ONLY_MULTIINSTANCE);
    
    return result;
  }

  /**
   * Returns the capabilities of this multi-instance classifier for the
   * relational data.
   *
   * @return            the capabilities of this object
   * @see               Capabilities
   */
  public Capabilities getMultiInstanceCapabilities() {
    Capabilities result = super.getCapabilities();
    
    // attributes
    result.enable(Capability.NOMINAL_ATTRIBUTES);
    result.enable(Capability.NUMERIC_ATTRIBUTES);
    result.enable(Capability.DATE_ATTRIBUTES);
    result.enable(Capability.MISSING_VALUES);

    // class
    result.disableAllClasses();
    result.enable(Capability.NO_CLASS);
    
    return result;
  }

  /**
   * Builds the classifier
   *
   * @param data the training data to be used for generating the
   * boosted classifier.
   * @throws Exception if the classifier could not be built successfully
   */
  public void buildClassifier(Instances data) throws Exception {
    // can classifier handle the data?
    getCapabilities().testWithFail(data);

    // remove instances with missing class
    Instances train = new Instances(data);
    train.deleteWithMissingClass();
    
    int numAttributes = train.attribute(1).relation().numAttributes();	
    m_Center = new double[numAttributes];

    if (getDebug())
      System.out.println("Start training ..."); 

    // convert the training dataset into single-instance dataset
    m_ConvertToSI.setInputFormat(train);	
    train = Filter.useFilter( train, m_ConvertToSI);

    if (m_filterType == FILTER_STANDARDIZE) 
      m_Filter = new Standardize();
    else if (m_filterType == FILTER_NORMALIZE)
      m_Filter = new Normalize();
    else 
      m_Filter = null;

    if (m_Filter!=null) {
      // normalize/standardize the converted training dataset
      m_Filter.setInputFormat(train);
      train = Filter.useFilter(train, m_Filter);
    }

    // convert the single-instance dataset into multi-instance dataset
    m_ConvertToMI.setInputFormat(train);
    train = Filter.useFilter(train, m_ConvertToMI);

    /*calculate all the distances (and store them in m_Distance[][][]), which
      are from each instance in all positive bags to all bags */
    calculateDistance(train);

    /*find the suitable ball center (m_Center) and the corresponding radius (m_Radius)*/
    findRadius(train); 

    if (getDebug())
      System.out.println("Finish building optimal ball model");
  }		



  /**