net.librec.eval.ranking.DiversityEvaluator Maven / Gradle / Ivy

Go to download
/**
 * Copyright (C) 2016 LibRec
 * 
 * This file is part of LibRec.
 * LibRec 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 3 of the License, or
 * (at your option) any later version.
 * 

 * LibRec 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 LibRec. If not, see .
 */
package net.librec.eval.ranking;

import net.librec.eval.AbstractRecommenderEvaluator;
import net.librec.math.structure.SparseMatrix;
import net.librec.math.structure.SymmMatrix;
import net.librec.recommender.item.ItemEntry;
import net.librec.recommender.item.RecommendedList;

import java.util.List;

/**
 * DiversityEvaluator, average dissimilarity of all pairs of items in the
 * recommended list at a specific cutoff position. Reference: Avoiding monotony:
 * improving the diversity of recommendation lists, ReSys, 2008
 *
 * @author WangYuFeng
 */
public class DiversityEvaluator extends AbstractRecommenderEvaluator {

    /**
     * Evaluate on the test set with the the list of recommended items.
     *
     * @param testMatrix
     *            the given test set
     * @param recommendedList
     *            the list of recommended items
     * @return evaluate result
     */
    public double evaluate(SparseMatrix testMatrix, RecommendedList recommendedList) {

        double totalDiversity = 0.0;
        int numUsers = testMatrix.numRows();
        int nonZeroNumUsers = 0;

        if (similarities.containsKey("item")) {
            SymmMatrix itemSimilarity = similarities.get("item").getSimilarityMatrix();
            for (int userID = 0; userID < numUsers; userID++) {
                List> recommendArrayListByUser = recommendedList.getItemIdxListByUserIdx(userID);
                if (recommendArrayListByUser.size() > 0) {
                    // calculate the sum of dissimilarities for each pair of items per user
                    double totalDisSimilarityPerUser = 0.0;
                    int topK = this.topN <= recommendArrayListByUser.size() ? this.topN : recommendArrayListByUser.size();
                    for (int i = 0; i < topK; i++) {
                        for (int j = 0; j < topK; j++) {
                            if (i == j) {
                                continue;
                            }
                            int item1 = recommendArrayListByUser.get(i).getKey();
                            int item2 = recommendArrayListByUser.get(j).getKey();
                            totalDisSimilarityPerUser += 1.0 - itemSimilarity.get(item1, item2);
                        }
                    }
                    totalDiversity += totalDisSimilarityPerUser * 2 / (topK * (topK - 1));
                    nonZeroNumUsers++;
                }
            }
        }

        return nonZeroNumUsers > 0 ? totalDiversity / nonZeroNumUsers : 0.0d;
    }
}