/*
* 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.
*/
/*
* MachineLearning09IBLR.java
* Copyright (C) 2009-2010 Aristotle University of Thessaloniki, Thessaloniki, Greece
*/
package mulan.experiments;
import java.util.ArrayList;
import java.util.List;
import java.util.Random;
import mulan.classifier.lazy.IBLR_ML;
import mulan.data.MultiLabelInstances;
import mulan.evaluation.Evaluation;
import mulan.evaluation.Evaluator;
import mulan.evaluation.MultipleEvaluation;
import mulan.evaluation.measure.AveragePrecision;
import mulan.evaluation.measure.Coverage;
import mulan.evaluation.measure.HammingLoss;
import mulan.evaluation.measure.Measure;
import mulan.evaluation.measure.OneError;
import mulan.evaluation.measure.RankingLoss;
import weka.core.Instances;
import weka.core.TechnicalInformation;
import weka.core.Utils;
import weka.core.TechnicalInformation.Field;
import weka.core.TechnicalInformation.Type;
/**
* Class replicating an experiment from a published paper
*
* @author Eleftherios Spyromitros-Xioufis (espyromi@csd.auth.gr)
* @version 2010.12.10
*/
public class MachineLearning09IBLR extends Experiment {
/**
* Main class
*
* @param args command line arguments
*/
public static void main(String[] args) {
try {
String path = Utils.getOption("path", args);
String filestem = Utils.getOption("filestem", args);
System.out.println("Loading the data set");
MultiLabelInstances dataSet = new MultiLabelInstances(path + filestem + ".arff", path + filestem + ".xml");
Evaluator evaluator = new Evaluator();
List<Measure> measures = new ArrayList<Measure>(5);
measures.add(new HammingLoss());
measures.add(new OneError());
measures.add(new Coverage());
measures.add(new RankingLoss());
measures.add(new AveragePrecision());
MultipleEvaluation iblrmlResults = new MultipleEvaluation();
MultipleEvaluation iblrmlPlusResults = new MultipleEvaluation();
Random random = new Random(1);
for (int repetition = 0; repetition < 10; repetition++) {
// perform 10-fold CV and add each to the current results
dataSet.getDataSet().randomize(random);
for (int fold = 0; fold < 10; fold++) {
System.out.println("Experiment " + (repetition * 10 + fold + 1));
Instances train = dataSet.getDataSet().trainCV(10, fold);
MultiLabelInstances multiTrain = new MultiLabelInstances(
train, dataSet.getLabelsMetaData());
Instances test = dataSet.getDataSet().testCV(10, fold);
MultiLabelInstances multiTest = new MultiLabelInstances(
test, dataSet.getLabelsMetaData());
System.out.println("IBLR-ML Experiment");
IBLR_ML iblrml = new IBLR_ML();
// iblrml.setDontNormalize(true);
iblrml.build(multiTrain);
evaluator = new Evaluator();
Evaluation e1 = evaluator.evaluate(iblrml, multiTest,
measures);
System.out.println(e1.toCSV());
iblrmlResults.addEvaluation(e1);
/*
* The following code produces the same results, as IBLR
* is equivalent to stacking using kNN at the 1st level
* and Logistic Regression at the 2nd level
*
* System.out.println("ML-Stacking Experiment");
* int numOfNeighbors = 10;
* Classifier baseClassifier = new IBk(numOfNeighbors);
* Classifier metaClassifier = new Logistic();
* MultiLabelStacking mls = new MultiLabelStacking( baseClassifier, metaClassifier);
* mls.setMetaPercentage(1.0);
* mls.build(multiTrain);
* evaluator = new Evaluator();
* Evaluation e1 = evaluator.evaluate(mls, multiTest, measures);
* System.out.println(e1.toCSV());
* iblrmlResults.addEvaluation(e1);
*/
System.out.println("IBLR-ML+ Experiment");
IBLR_ML iblrmlplus = new IBLR_ML();
iblrmlplus.setAddFeatures(true);
// iblrmlplus.setDontNormalize(true);
iblrmlplus.build(multiTrain);
evaluator = new Evaluator();
Evaluation e2 = evaluator.evaluate(iblrmlplus, multiTest,
measures);
System.out.println(e2.toCSV());
iblrmlPlusResults.addEvaluation(e2);
}
}
iblrmlResults.calculateStatistics();
System.out.println(iblrmlResults);
iblrmlPlusResults.calculateStatistics();
System.out.println(iblrmlPlusResults);
} catch (Exception e) {
e.printStackTrace();
}
}
@Override
public TechnicalInformation getTechnicalInformation() {
TechnicalInformation result;
result = new TechnicalInformation(Type.ARTICLE);
result.setValue(Field.AUTHOR, "Weiwei Cheng and Eyke Hullermeier");
result.setValue(Field.TITLE, "Combining instance-based learning and logistic regression for multilabel classification ");
result.setValue(Field.JOURNAL, "Machine Learning");
result.setValue(Field.VOLUME, "76");
result.setValue(Field.NUMBER, "2-3");
result.setValue(Field.YEAR, "2009");
result.setValue(Field.ISSN, "0885-6125");
result.setValue(Field.PAGES, "211-225");
result.setValue(Field.PUBLISHER, "Springer Netherlands");
return result;
}
}