/*
* 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.
*/
/*
* DedupingEvaluation.java
* Copyright (C) 2003 Mikhail Bilenko
*
*/
package weka.deduping;
import java.util.*;
import java.io.*;
import weka.core.*;
import weka.filters.Filter;
import weka.filters.unsupervised.attribute.Remove;
/**
* Class for evaluating deduping
*
* @author Mikhail Bilenko
*/
public class DedupingEvaluation {
/** The number of produced clusters */
protected int m_numClusters;
/** Training instances */
protected Instances m_trainInstances;
/** Test instances */
protected Instances m_testInstances;
/** Array for storing the confusion matrix. */
protected double [][] m_ConfusionMatrix;
/**
* Returns a string describing this evaluator
* @return a description of the evaluator suitable for
* displaying in the explorer/experimenter gui
*/
public String globalInfo() {
return " A deduping evaluator that evaluates results of running a "
+ "deduping experiment.";
}
/** A default constructor */
public DedupingEvaluation () {
}
/** Train a deduper on the supplied data
* @param deduper a deduper to train
* @param data training data
*/
public void trainDeduper(Deduper deduper, Instances trainingData, Instances testData) throws Exception {
deduper.buildDeduper(trainingData, testData);
}
/**
* Evaluates the deduper on a given set of test instances
*
* @param clusterer semi-supervised clusterer
* @param testInstances set of test instances for evaluation
* @return a list of arrays containing the basic statistics for each point
* @exception Exception if model could not be evaluated successfully
*/
public ArrayList evaluateModel (Deduper deduper, Instances testInstances) throws Exception {
ArrayList resultList = new ArrayList();
m_testInstances = testInstances;
// Run the deduper collecting data
int numTrueClasses = countPresentClasses(testInstances);
int numObjects = (int) (0.8 * numTrueClasses);
System.out.println("testInstances: " + testInstances.numInstances() + " true=" + numTrueClasses + " desired:" + numObjects);
System.out.println("numClasses=" + testInstances.numClasses());
deduper.findDuplicates(testInstances, numObjects);
return deduper.getStatistics();
}
/** A helper function that determines how many classes are actually
* represented in an Instances object
* @param instances a set of instances
* @return the number of classes present among the instances
*/
protected int countPresentClasses(Instances instances) {
HashSet classValueSet = new HashSet();
for (int i = 0; i < instances.numInstances(); i++) {
Instance instance = (Instance) instances.instance(i);
classValueSet.add(new Double(instance.classValue()));
}
return classValueSet.size();
}
}