/*
* 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.
*/
/*
* RankingLoss.java
* Copyright (C) 2009-2010 Aristotle University of Thessaloniki, Thessaloniki, Greece
*/
package mulan.evaluation.loss;
import java.util.ArrayList;
/**
* Implementation of the "ranking loss" ranking loss function. It is basically
* the size of the error set divided by all possible pairs of relevant and
* irrelevant labels
*
* @author Grigorios Tsoumakas
* @version 2010.11.05
*/
public class RankingLoss extends ErrorSetSize {
@Override
public String getName() {
return "Ranking Loss";
}
@Override
public double computeLoss(int[] ranking, boolean[] groundTruth) {
int numLabels = groundTruth.length;
ArrayList<Integer> trueIndexes = new ArrayList<Integer>();
ArrayList<Integer> falseIndexes = new ArrayList<Integer>();
for (int labelIndex = 0; labelIndex < numLabels; labelIndex++) {
if (groundTruth[labelIndex]) {
trueIndexes.add(labelIndex);
} else {
falseIndexes.add(labelIndex);
}
}
if (trueIndexes.size() != 0 && falseIndexes.size() != 0) {
int rolp = 0; // reversed ordered label pairs
for (int k : trueIndexes) {
for (int l : falseIndexes) {
// if (output[instanceIndex].getConfidences()[trueIndexes.get(k)] <= output[instanceIndex].getConfidences()[falseIndexes.get(l)])
if (ranking[k] > ranking[l]) {
rolp++;
}
}
}
return (double) rolp / (trueIndexes.size() * falseIndexes.size());
} else {
return 0;
}
}
}