oner.java

Java 编写的多种数据挖掘算法 包括聚类、分类、预处理等

/*
 *    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.
 */

/*
 *    OneR.java
 *    Copyright (C) 1999 Ian H. Witten
 *
 */

package weka.classifiers.rules;

import weka.classifiers.Classifier;
import weka.classifiers.Evaluation;
import weka.core.Attribute;
import weka.core.Capabilities;
import weka.core.Instance;
import weka.core.Instances;
import weka.core.Option;
import weka.core.OptionHandler;
import weka.core.Utils;
import weka.core.TechnicalInformation;
import weka.core.TechnicalInformation.Type;
import weka.core.TechnicalInformation.Field;
import weka.core.TechnicalInformationHandler;
import weka.core.Capabilities.Capability;

import java.io.Serializable;
import java.util.Enumeration;
import java.util.Vector;

/**
 <!-- globalinfo-start -->
 * Class for building and using a 1R classifier; in other words, uses the minimum-error attribute for prediction, discretizing numeric attributes. For more information, see:<br/>
 * <br/>
 * R.C. Holte (1993). Very simple classification rules perform well on most commonly used datasets. Machine Learning. 11:63-91.
 * <p/>
 <!-- globalinfo-end -->
 *
 <!-- technical-bibtex-start -->
 * BibTeX:
 * <pre>
 * &#64;article{Holte1993,
 *    author = {R.C. Holte},
 *    journal = {Machine Learning},
 *    pages = {63-91},
 *    title = {Very simple classification rules perform well on most commonly used datasets},
 *    volume = {11},
 *    year = {1993}
 * }
 * </pre>
 * <p/>
 <!-- technical-bibtex-end -->
 *
 <!-- options-start -->
 * Valid options are: <p/>
 * 
 * <pre> -B &lt;minimum bucket size&gt;
 *  The minimum number of objects in a bucket (default: 6).</pre>
 * 
 <!-- options-end -->
 * 
 * @author Ian H. Witten ([e-mail])
 * @version $Revision: 1.20 $ 
*/
public class OneR 
  extends Classifier 
  implements OptionHandler, TechnicalInformationHandler {
    
  /** for serialization */
  static final long serialVersionUID = -2459427002147861445L;
  
  /**
   * Returns a string describing classifier
   * @return a description suitable for
   * displaying in the explorer/experimenter gui
   */
  public String globalInfo() {

    return "Class for building and using a 1R classifier; in other words, uses "
      + "the minimum-error attribute for prediction, discretizing numeric "
      + "attributes. 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;
    
    result = new TechnicalInformation(Type.ARTICLE);
    result.setValue(Field.AUTHOR, "R.C. Holte");
    result.setValue(Field.YEAR, "1993");
    result.setValue(Field.TITLE, "Very simple classification rules perform well on most commonly used datasets");
    result.setValue(Field.JOURNAL, "Machine Learning");
    result.setValue(Field.VOLUME, "11");
    result.setValue(Field.PAGES, "63-91");
    
    return result;
  }

  /**
   * Class for storing store a 1R rule.
   */
  private class OneRRule 
    implements Serializable {
    
    /** for serialization */
    static final long serialVersionUID = 1152814630957092281L;

    /** The class attribute. */
    private Attribute m_class;

    /** The number of instances used for building the rule. */
    private int m_numInst;

    /** Attribute to test */
    private Attribute m_attr; 

    /** Training set examples this rule gets right */
    private int m_correct; 

    /** Predicted class for each value of attr */
    private int[] m_classifications; 

    /** Predicted class for missing values */
    private int m_missingValueClass = -1; 

    /** Breakpoints (numeric attributes only) */
    private double[] m_breakpoints; 
  
    /**
     * Constructor for nominal attribute.
     * 
     * @param data the data to work with
     * @param attribute the attribute to use
     * @throws Exception if something goes wrong
     */
    public OneRRule(Instances data, Attribute attribute) throws Exception {

      m_class = data.classAttribute();
      m_numInst = data.numInstances();
      m_attr = attribute;
      m_correct = 0;
      m_classifications = new int[m_attr.numValues()];
    }

    /**
     * Constructor for numeric attribute.
     * 
     * @param data the data to work with
     * @param attribute the attribute to use
     * @param nBreaks the break point
     * @throws Exception if something goes wrong
     */
    public OneRRule(Instances data, Attribute attribute, int nBreaks) throws Exception {

      m_class = data.classAttribute();
      m_numInst = data.numInstances();
      m_attr = attribute;
      m_correct = 0;
      m_classifications = new int[nBreaks];
      m_breakpoints = new double[nBreaks - 1]; // last breakpoint is infinity
    }
    
    /**
     * Returns a description of the rule.
     * 
     * @return a string representation of the rule
     */
    public String toString() {

      try {
	StringBuffer text = new StringBuffer();
	text.append(m_attr.name() + ":\n");
	for (int v = 0; v < m_classifications.length; v++) {
	  text.append("\t");
	  if (m_attr.isNominal()) {
	    text.append(m_attr.value(v));
	  } else if (v < m_breakpoints.length) {
	    text.append("< " + m_breakpoints[v]);
	  } else if (v > 0) {
	    text.append(">= " + m_breakpoints[v - 1]);
	  } else {
	    text.append("not ?");
	  }
	  text.append("\t-> " + m_class.value(m_classifications[v]) + "\n");
	}
	if (m_missingValueClass != -1) {
	  text.append("\t?\t-> " + m_class.value(m_missingValueClass) + "\n");
	}
	text.append("(" + m_correct + "/" + m_numInst + " instances correct)\n");
	return text.toString();
      } catch (Exception e) {
	return "Can't print OneR classifier!";
      }
    }
  }
  
  /** A 1-R rule */
  private OneRRule m_rule;

  /** The minimum bucket size */
  private int m_minBucketSize = 6;

  /**
   * Classifies a given instance.
   *
   * @param inst the instance to be classified
   * @return the classification of the instance
   */
  public double classifyInstance(Instance inst) {

    int v = 0;
    if (inst.isMissing(m_rule.m_attr)) {
      if (m_rule.m_missingValueClass != -1) {
	return m_rule.m_missingValueClass;
      } else {
	return 0;  // missing values occur in test but not training set    
      }
    }
    if (m_rule.m_attr.isNominal()) {
      v = (int) inst.value(m_rule.m_attr);
    } else {
      while (v < m_rule.m_breakpoints.length &&
	     inst.value(m_rule.m_attr) >= m_rule.m_breakpoints[v]) {
	v++;
      }
    }
    return m_rule.m_classifications[v];
  }

  /**
   * 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.NUMERIC_ATTRIBUTES);
    result.enable(Capability.DATE_ATTRIBUTES);
    result.enable(Capability.MISSING_VALUES);

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

    return result;
  }

  /**
   * Generates the classifier.
   *
   * @param instances the instances to be used for building the classifier
   * @throws Exception if the classifier can't be built successfully
   */
  public void buildClassifier(Instances instances) 
    throws Exception {
    
    boolean noRule = true;

    // can classifier handle the data?
    getCapabilities().testWithFail(instances);

    // remove instances with missing class
    Instances data = new Instances(instances);