Example usage for org.apache.mahout.cf.taste.impl.eval RMSRecommenderEvaluator RMSRecommenderEvaluator

Introduction

In this page you can find the example usage for org.apache.mahout.cf.taste.impl.eval RMSRecommenderEvaluator RMSRecommenderEvaluator.

Prototype

RMSRecommenderEvaluator

Source Link

Usage

From source file:hr.fer.tel.rovkp.homework03.task02.ItemBasedRecommenderEvaluator.java

public static void main(String[] args) throws Exception {

    RandomUtils.useTestSeed();/*from w  w w  .  j  a v  a2  s.  co m*/

    String fileItemsSimilarity = "./src/main/resources/item_similarity.csv";
    String fileDataModel = "./src/main/resources/jester_ratings_small.dat";

    DataModel model = new FileDataModel(new File(fileDataModel));

    RecommenderBuilder builder = (DataModel m) -> RecommenderUtils.itemBasedRecommender(m, fileItemsSimilarity);

    RecommenderEvaluator recEvaluator = new RMSRecommenderEvaluator();
    double score = recEvaluator.evaluate(builder, null, model, 0.7, 1.0);
    System.out.println(score);
}

---

From source file:hr.fer.tel.rovkp.homework03.task02.UserBasedRecommenderEvaluator.java

public static void main(String[] args) throws Exception {

    RandomUtils.useTestSeed();/* w  w w .  j a  v  a2 s  .  co  m*/

    String fileDataModel = "./src/main/resources/jester_ratings.dat";

    DataModel model = new FileDataModel(new File(fileDataModel));

    RecommenderBuilder builder = (DataModel m) -> RecommenderUtils.userBasedRecommender(m);

    RecommenderEvaluator recEvaluator = new RMSRecommenderEvaluator();
    double score = recEvaluator.evaluate(builder, null, model, 0.7, 1.0);
    System.out.println(score);
}

---

From source file:hr.fer.tel.rovkp.lab03.HybridRecommenderEvaluator.java

public static void main(String[] args) throws IOException, TasteException {
    DataModel model = new FileDataModel(new File("./src/main/resources/jester_ratings.dat"), ",");
    System.out.println(new RMSRecommenderEvaluator().evaluate(hybridRecommender(), null, model, 0.7, 1.0));
}

---

From source file:net.aprendizajengrande.gitrecommender.Experiment.java

License:Open Source License

public static void main(String[] args) throws Exception {

    DataModel model = new FileDataModel(new File(args[0]));

    System.out.println("For repo: " + args[0]);
    System.out.println("User-based:");

    RecommenderEvaluator evaluator = new RMSRecommenderEvaluator();
    RecommenderBuilder recommenderBuilder = new RecommenderBuilder() {
        @Override/*from w w w  .j  a  v a2s  . c  om*/
        public Recommender buildRecommender(DataModel model) throws TasteException {
            UserSimilarity similarity = new PearsonCorrelationSimilarity(model);
            UserNeighborhood neighborhood = new NearestNUserNeighborhood(2, similarity, model);
            return new GenericUserBasedRecommender(model, neighborhood, similarity);
        }
    };
    // Use 90% of the data to train; test using the other 10%.
    double score = evaluator.evaluate(recommenderBuilder, null, model, 0.9, 1.0);
    System.out.println(score);

    recommenderBuilder = new RecommenderBuilder() {
        @Override
        public Recommender buildRecommender(DataModel model) throws TasteException {
            ItemSimilarity similarity = new EuclideanDistanceSimilarity(model);
            //new LogLikelihoodSimilarity(model);
            return new GenericItemBasedRecommender(model, similarity);
        }
    };
    System.out.println("Item-based:");
    score = evaluator.evaluate(recommenderBuilder, null, model, 0.9, 1.0);
    System.out.println(score);
}

---

From source file:nl.gridline.zieook.client.tools.ZieOokEvaluatorTest.java

License:Apache License

/**
 * 3x userbased using RMSRecommenderEvaluator
 *//*from  www . j a  v  a  2 s.c o  m*/
@Test
@Ignore
public void evaulate2() {
    // UserBasedRecommenderBuilder
    // RMSRecommenderEvaluator
    try {
        DataModel model = createDataModel(input);

        RecommenderEvaluator evaluator = new RMSRecommenderEvaluator();
        double evaluation = evaluator.evaluate(
                new UserBasedRecommenderBuilder(EuclideanDistanceSimilarity.class.getCanonicalName()), null,
                model, 0.9, 1.0);
        LOG.info("result: " + evaluation);
        writetofile("UserBasedRecommenderBuilder,EuclideanDistanceSimilarity,RMSRecommenderEvaluator,"
                + evaluation + "\n");

        evaluator = new RMSRecommenderEvaluator();
        evaluation = evaluator.evaluate(
                new UserBasedRecommenderBuilder(PearsonCorrelationSimilarity.class.getCanonicalName()), null,
                model, 0.9, 1.0);
        LOG.info("result: " + evaluation);
        writetofile("UserBasedRecommenderBuilder,PearsonCorrelationSimilarity,RMSRecommenderEvaluator,"
                + evaluation + "\n");

        evaluator = new RMSRecommenderEvaluator();
        evaluation = evaluator.evaluate(
                new UserBasedRecommenderBuilder(UncenteredCosineSimilarity.class.getCanonicalName()), null,
                model, 0.9, 1.0);
        LOG.info("result: " + evaluation);
        writetofile("UserBasedRecommenderBuilder,UncenteredCosineSimilarity,RMSRecommenderEvaluator,"
                + evaluation + "\n");

    } catch (TasteException e) {
        LOG.error("faild evaulate", e);
        fail();
    }
}

---

From source file:nl.gridline.zieook.client.tools.ZieOokEvaluatorTest.java

License:Apache License

@Test
@Ignore/* w ww .  j ava2 s . co m*/
public void evaluate3() {
    // ItemBasedRecommenderBuilder
    // RMSRecommenderEvaluator
    try {
        DataModel model = createDataModel(input);
        RecommenderEvaluator evaluator = new RMSRecommenderEvaluator();
        double evaluation = evaluator.evaluate(
                new ItemBasedRecommenderBuilder(EuclideanDistanceSimilarity.class.getCanonicalName()), null,
                model, 0.9, 1.0);
        LOG.info("result: " + evaluation);
        writetofile(
                "ItemBasedRecommenderBuilder,EuclideanDistanceSimilarity,AverageAbsoluteDifferenceRecommenderEvaluator,"
                        + evaluation + "\n");

        evaluator = new RMSRecommenderEvaluator();
        evaluation = evaluator.evaluate(
                new ItemBasedRecommenderBuilder(PearsonCorrelationSimilarity.class.getCanonicalName()), null,
                model, 0.9, 1.0);
        LOG.info("result: " + evaluation);
        writetofile(
                "ItemBasedRecommenderBuilder,PearsonCorrelationSimilarity,AverageAbsoluteDifferenceRecommenderEvaluator,"
                        + evaluation + "\n");

        evaluator = new RMSRecommenderEvaluator();
        evaluation = evaluator.evaluate(
                new ItemBasedRecommenderBuilder(UncenteredCosineSimilarity.class.getCanonicalName()), null,
                model, 0.9, 1.0);
        LOG.info("result: " + evaluation);
        writetofile(
                "ItemBasedRecommenderBuilder,UncenteredCosineSimilarity,AverageAbsoluteDifferenceRecommenderEvaluator,"
                        + evaluation + "\n");
    } catch (TasteException e) {
        LOG.error("faild evaulate", e);
        fail();
    }
}

---

From source file:nl.gridline.zieook.tasks.quality.MahoutEvaluatorTest.java

License:Apache License

@Test
public void testRandom() throws IOException, TasteException {
    RandomUtils.useTestSeed();//ww  w  .j a  v a 2 s.c  o m

    LOG.info("testing Random:");
    DataModel model = new FileDataModel(testData);
    GenericRecommenderIRStatsEvaluator evaluatorIR = new GenericRecommenderIRStatsEvaluator();
    AverageAbsoluteDifferenceRecommenderEvaluator evaluator = new AverageAbsoluteDifferenceRecommenderEvaluator();
    RMSRecommenderEvaluator rmsEvaluator = new RMSRecommenderEvaluator();

    RecommenderBuilder builder = new RecommenderBuilder() {
        @Override
        public Recommender buildRecommender(DataModel model) throws TasteException {
            return new RandomRecommender(model);
        }
    };

    long start = System.currentTimeMillis();
    double score = evaluator.evaluate(builder, null, model, 0.7, 1.0);
    LOG.info("score: {} in {}s", String.format("%.4f", score), ((System.currentTimeMillis() - start) / 1000));

    start = System.currentTimeMillis();
    score = rmsEvaluator.evaluate(builder, null, model, 0.7, 1.0);
    LOG.info("rms score: {} in {}s", String.format("%.4f", score),
            ((System.currentTimeMillis() - start) / 1000));

    start = System.currentTimeMillis();
    IRStatistics stats = evaluatorIR.evaluate(builder, null, model, null, 2,
            GenericRecommenderIRStatsEvaluator.CHOOSE_THRESHOLD, 1.0);
    LOG.info("precision: {} recall: {} in {}s", new Object[] { String.format("%.4f", stats.getPrecision()),
            String.format("%.4f", stats.getRecall()), ((System.currentTimeMillis() - start) / 1000) });

}

---

From source file:nl.gridline.zieook.tasks.quality.MahoutEvaluatorTest.java

License:Apache License

@Test
public void testPearsonCorrelationSimilarity() throws IOException, TasteException {
    RandomUtils.useTestSeed();/* w w  w . j  av a 2  s  .co  m*/

    LOG.info("testing PearsonCorrelationSimilarity: ");

    DataModel model = new FileDataModel(testData);

    AverageAbsoluteDifferenceRecommenderEvaluator evaluator = new AverageAbsoluteDifferenceRecommenderEvaluator();
    RMSRecommenderEvaluator rmsEvaluator = new RMSRecommenderEvaluator();

    GenericRecommenderIRStatsEvaluator evaluatorIR = new GenericRecommenderIRStatsEvaluator();

    RecommenderBuilder builder = new RecommenderBuilder() {
        @Override
        public Recommender buildRecommender(DataModel model) throws TasteException {
            PearsonCorrelationSimilarity similarity = new PearsonCorrelationSimilarity(model);
            UserNeighborhood neighborhood = new NearestNUserNeighborhood(2, similarity, model);
            return new GenericUserBasedRecommender(model, neighborhood, similarity);
        }
    };

    long start = System.currentTimeMillis();
    double score = evaluator.evaluate(builder, null, model, 0.7, 1.0);
    LOG.info("score: {} in {}s", String.format("%.4f", score), ((System.currentTimeMillis() - start) / 1000));

    start = System.currentTimeMillis();
    score = rmsEvaluator.evaluate(builder, null, model, 0.7, 1.0);
    LOG.info("rms score: {} in {}s", String.format("%.4f", score),
            ((System.currentTimeMillis() - start) / 1000));

    start = System.currentTimeMillis();
    IRStatistics stats = evaluatorIR.evaluate(builder, null, model, null, 2,
            GenericRecommenderIRStatsEvaluator.CHOOSE_THRESHOLD, 1.0);

    LOG.info("precision: {} recall: {} in {}s", new Object[] { String.format("%.4f", stats.getPrecision()),
            String.format("%.4f", stats.getRecall()), ((System.currentTimeMillis() - start) / 1000) });

}

---

From source file:nl.gridline.zieook.tasks.quality.MahoutEvaluatorTest.java

License:Apache License

@Test
public void testTanimotoCoefficientSimilarity() throws TasteException, IOException {
    RandomUtils.useTestSeed();/*from  www  .  j  a va 2s . c om*/

    LOG.info("testing: TanimotoCoefficientSimilarity:");

    DataModel model = new FileDataModel(testData);

    AverageAbsoluteDifferenceRecommenderEvaluator evaluator = new AverageAbsoluteDifferenceRecommenderEvaluator();
    RMSRecommenderEvaluator rmsEvaluator = new RMSRecommenderEvaluator();

    GenericRecommenderIRStatsEvaluator evaluatorIR = new GenericRecommenderIRStatsEvaluator();

    RecommenderBuilder builder = new RecommenderBuilder() {
        @Override
        public Recommender buildRecommender(DataModel model) throws TasteException {
            TanimotoCoefficientSimilarity similarity = new TanimotoCoefficientSimilarity(model);
            UserNeighborhood neighborhood = new NearestNUserNeighborhood(2, similarity, model);
            return new GenericUserBasedRecommender(model, neighborhood, similarity);
        }
    };

    long start = System.currentTimeMillis();
    double score = evaluator.evaluate(builder, null, model, 0.7, 1.0);
    LOG.info("average score: {} in {}s", String.format("%.4f", score),
            ((System.currentTimeMillis() - start) / 1000));

    start = System.currentTimeMillis();
    score = rmsEvaluator.evaluate(builder, null, model, 0.7, 1.0);
    LOG.info("rms score: {} in {}s", String.format("%.4f", score),
            ((System.currentTimeMillis() - start) / 1000));

    start = System.currentTimeMillis();
    IRStatistics stats = evaluatorIR.evaluate(builder, null, model, null, 2,
            GenericRecommenderIRStatsEvaluator.CHOOSE_THRESHOLD, 1.0);

    LOG.info("precision: {} recall: {} in {}s", new Object[] { String.format("%.4f", stats.getPrecision()),
            String.format("%.4f", stats.getRecall()), ((System.currentTimeMillis() - start) / 1000) });
}

---

From source file:nl.gridline.zieook.tasks.quality.MahoutEvaluatorTest.java

License:Apache License

@Test
public void testUncenteredCosineSimilarity() throws TasteException, IOException {
    RandomUtils.useTestSeed();/*from www.  j a v a  2  s .  c o  m*/

    LOG.info("testing: UncenteredCosineSimilarity:");

    DataModel model = new FileDataModel(testData);

    AverageAbsoluteDifferenceRecommenderEvaluator evaluator = new AverageAbsoluteDifferenceRecommenderEvaluator();
    RMSRecommenderEvaluator rmsEvaluator = new RMSRecommenderEvaluator();

    GenericRecommenderIRStatsEvaluator evaluatorIR = new GenericRecommenderIRStatsEvaluator();

    RecommenderBuilder builder = new RecommenderBuilder() {
        @Override
        public Recommender buildRecommender(DataModel model) throws TasteException {
            UncenteredCosineSimilarity similarity = new UncenteredCosineSimilarity(model);
            UserNeighborhood neighborhood = new NearestNUserNeighborhood(2, similarity, model);
            return new GenericUserBasedRecommender(model, neighborhood, similarity);
        }
    };

    long start = System.currentTimeMillis();
    double score = evaluator.evaluate(builder, null, model, 0.7, 1.0);
    LOG.info("average score: {} in {}s", String.format("%.4f", score),
            ((System.currentTimeMillis() - start) / 1000));

    start = System.currentTimeMillis();
    score = rmsEvaluator.evaluate(builder, null, model, 0.7, 1.0);
    LOG.info("rms score: {} in {}s", String.format("%.4f", score),
            ((System.currentTimeMillis() - start) / 1000));

    start = System.currentTimeMillis();
    IRStatistics stats = evaluatorIR.evaluate(builder, null, model, null, 2,
            GenericRecommenderIRStatsEvaluator.CHOOSE_THRESHOLD, 1.0);

    LOG.info("precision: {} recall: {} in {}s", new Object[] { String.format("%.4f", stats.getPrecision()),
            String.format("%.4f", stats.getRecall()), ((System.currentTimeMillis() - start) / 1000) });
}

---

• «
• 1
• 2
• »