Package org.apache.mahout.cf.taste.recommender

Examples of org.apache.mahout.cf.taste.recommender.Recommender


/** <p>Tests {@link GenericUserBasedRecommender}.</p> */
public final class GenericUserBasedRecommenderTest extends TasteTestCase {

  @Test
  public void testRecommender() throws Exception {
    Recommender recommender = buildRecommender();
    List<RecommendedItem> recommended = recommender.recommend(1, 1);
    assertNotNull(recommended);
    assertEquals(1, recommended.size());
    RecommendedItem firstRecommended = recommended.get(0);
    assertEquals(2, firstRecommended.getItemID());
    assertEquals(0.1f, firstRecommended.getValue(), EPSILON);
    recommender.refresh(null);
    assertEquals(2, firstRecommended.getItemID());
    assertEquals(0.1f, firstRecommended.getValue(), EPSILON);
  }
View Full Code Here


                    {0.1, 0.4, 0.5, 0.8, 0.9, 1.0},
                    {0.2, 0.3, 0.6, 0.7, 0.1, 0.2},
            });
    UserSimilarity similarity = new PearsonCorrelationSimilarity(dataModel);
    UserNeighborhood neighborhood = new NearestNUserNeighborhood(2, similarity, dataModel);
    Recommender recommender = new GenericUserBasedRecommender(dataModel, neighborhood, similarity);
    List<RecommendedItem> fewRecommended = recommender.recommend(1, 2);
    List<RecommendedItem> moreRecommended = recommender.recommend(1, 4);
    for (int i = 0; i < fewRecommended.size(); i++) {
      assertEquals(fewRecommended.get(i).getItemID(), moreRecommended.get(i).getItemID());
    }
    recommender.refresh(null);
    for (int i = 0; i < fewRecommended.size(); i++) {
      assertEquals(fewRecommended.get(i).getItemID(), moreRecommended.get(i).getItemID());
    }
  }
View Full Code Here

                    {0.2, 0.3, 0.3, 0.6},
                    {0.4, 0.5, 0.5, 0.9},
            });
    UserSimilarity similarity = new PearsonCorrelationSimilarity(dataModel);
    UserNeighborhood neighborhood = new NearestNUserNeighborhood(2, similarity, dataModel);
    Recommender recommender = new GenericUserBasedRecommender(dataModel, neighborhood, similarity);
    List<RecommendedItem> originalRecommended = recommender.recommend(1, 2);
    List<RecommendedItem> rescoredRecommended =
        recommender.recommend(1, 2, new ReversingRescorer<Long>());
    assertNotNull(originalRecommended);
    assertNotNull(rescoredRecommended);
    assertEquals(2, originalRecommended.size());
    assertEquals(2, rescoredRecommended.size());
    assertEquals(originalRecommended.get(0).getItemID(), rescoredRecommended.get(1).getItemID());
View Full Code Here

    assertEquals(originalRecommended.get(1).getItemID(), rescoredRecommended.get(0).getItemID());
  }

  @Test
  public void testEstimatePref() throws Exception {
    Recommender recommender = buildRecommender();
    assertEquals(0.1f, recommender.estimatePreference(1, 2), EPSILON);
  }
View Full Code Here

    assertEquals(0.1f, recommender.estimatePreference(1, 2), EPSILON);
  }

  @Test
  public void testBestRating() throws Exception {
    Recommender recommender = buildRecommender();
    List<RecommendedItem> recommended = recommender.recommend(1, 1);
    assertNotNull(recommended);
    assertEquals(1, recommended.size());
    RecommendedItem firstRecommended = recommended.get(0);
    // item one should be recommended because it has a greater rating/score
    assertEquals(2, firstRecommended.getItemID());
View Full Code Here

      howMany = Integer.parseInt(args[1]);
    }

    System.out.println("Run Items");
    ItemSimilarity similarity = new EuclideanDistanceSimilarity(model);
    Recommender recommender = new GenericItemBasedRecommender(model, similarity); // Use an item-item recommender
    for (int i = 0; i < LOOPS; i++){
      LoadStatistics loadStats = LoadEvaluator.runLoad(recommender, howMany);
      System.out.println(loadStats);
    }
View Full Code Here

  @Test
  public void testFile() throws Exception {
    UserSimilarity userSimilarity = new PearsonCorrelationSimilarity(model);
    UserNeighborhood neighborhood = new NearestNUserNeighborhood(3, userSimilarity, model);
    Recommender recommender = new GenericUserBasedRecommender(model, neighborhood, userSimilarity);
    assertEquals(1, recommender.recommend(123, 3).size());
    assertEquals(0, recommender.recommend(234, 3).size());
    assertEquals(1, recommender.recommend(345, 3).size());

    // Make sure this doesn't throw an exception
    model.refresh(null);
  }
View Full Code Here

public final class CachingRecommenderTest extends TasteTestCase {

  @Test
  public void testRecommender() throws Exception {
    MutableInt recommendCount = new MutableInt();
    Recommender mockRecommender = new MockRecommender(recommendCount);

    Recommender cachingRecommender = new CachingRecommender(mockRecommender);
    cachingRecommender.recommend(1, 1);
    assertEquals(1, recommendCount.intValue());
    cachingRecommender.recommend(2, 1);
    assertEquals(2, recommendCount.intValue());
    cachingRecommender.recommend(1, 1);
    assertEquals(2, recommendCount.intValue());
    cachingRecommender.recommend(2, 1);
    assertEquals(2, recommendCount.intValue());
    cachingRecommender.refresh(null);
    cachingRecommender.recommend(1, 1);
    assertEquals(3, recommendCount.intValue());
    cachingRecommender.recommend(2, 1);
    assertEquals(4, recommendCount.intValue());
    cachingRecommender.recommend(3, 1);
    assertEquals(5, recommendCount.intValue());

    // Results from this recommend() method can be cached...
    IDRescorer rescorer = NullRescorer.getItemInstance();
    cachingRecommender.refresh(null);
    cachingRecommender.recommend(1, 1, rescorer);
    assertEquals(6, recommendCount.intValue());
    cachingRecommender.recommend(2, 1, rescorer);
    assertEquals(7, recommendCount.intValue());
    cachingRecommender.recommend(1, 1, rescorer);
    assertEquals(7, recommendCount.intValue());
    cachingRecommender.recommend(2, 1, rescorer);
    assertEquals(7, recommendCount.intValue());

    // until you switch Rescorers
    cachingRecommender.recommend(1, 1, null);
    assertEquals(8, recommendCount.intValue());
    cachingRecommender.recommend(2, 1, null);
    assertEquals(9, recommendCount.intValue());

    cachingRecommender.refresh(null);
    cachingRecommender.estimatePreference(1, 1);
    assertEquals(10, recommendCount.intValue());
    cachingRecommender.estimatePreference(1, 2);
    assertEquals(11, recommendCount.intValue());
    cachingRecommender.estimatePreference(1, 2);
    assertEquals(11, recommendCount.intValue());
  }
View Full Code Here

public final class KnnItemBasedRecommenderTest extends TasteTestCase {

  @Test
  public void testRecommender() throws Exception {
    Recommender recommender = buildRecommender();
    List<RecommendedItem> recommended = recommender.recommend(1, 1);
    assertNotNull(recommended);
    assertEquals(1, recommended.size());
    RecommendedItem firstRecommended = recommended.get(0);
    assertEquals(2, firstRecommended.getItemID());
    assertEquals(0.1f, firstRecommended.getValue(), EPSILON);
    recommender.refresh(null);
    assertEquals(2, firstRecommended.getItemID());
    assertEquals(0.1f, firstRecommended.getValue(), EPSILON);
  }
View Full Code Here

                    {0.1, 0.4, 0.5, 0.8, 0.9, 1.0},
                    {0.2, 0.3, 0.6, 0.7, 0.1, 0.2},
            });
    ItemSimilarity similarity = new PearsonCorrelationSimilarity(dataModel);
    Optimizer optimizer = new ConjugateGradientOptimizer();
    Recommender recommender = new KnnItemBasedRecommender(dataModel, similarity, optimizer, 5);
    List<RecommendedItem> fewRecommended = recommender.recommend(1, 2);
    List<RecommendedItem> moreRecommended = recommender.recommend(1, 4);
    for (int i = 0; i < fewRecommended.size(); i++) {
      assertEquals(fewRecommended.get(i).getItemID(), moreRecommended.get(i).getItemID());
    }
    recommender.refresh(null);
    for (int i = 0; i < fewRecommended.size(); i++) {
      assertEquals(fewRecommended.get(i).getItemID(), moreRecommended.get(i).getItemID());
    }
  }
View Full Code Here

TOP

Related Classes of org.apache.mahout.cf.taste.recommender.Recommender

Copyright © 2018 www.massapicom. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.