Example usage for org.apache.mahout.cf.taste.eval IRStatistics getF1Measure

Introduction

In this page you can find the example usage for org.apache.mahout.cf.taste.eval IRStatistics getF1Measure.

Prototype

double getF1Measure();

Source Link

Document

<p> See <a href="http://en.wikipedia.org/wiki/Information_retrieval#F-measure">F-measure</a>.

Usage

From source file:edu.uci.ics.sourcerer.ml.db.tools.RecommenderEvaluatorTest.java

License:Open Source License

public void testEvaluate() throws Exception {
    DataModel model = getDataModel();/*from w  w w . ja  va 2  s  .c  o m*/

    RecommenderBuilder builder = new RecommenderBuilder() {
        @Override
        public Recommender buildRecommender(DataModel dataModel) throws TasteException {

            // return new SlopeOneRecommender(dataModel);

            // return new SVDRecommender(dataModel, 25, 100);

            //return new RandomRecommender(dataModel);

            return new ApiRecommender(dataModel);

            //            return new TreeClusteringRecommender(
            //                  dataModel,
            //                  new NearestNeighborClusterSimilarity(
            //                        //UserSimilarity similarity = new EuclideanDistanceSimilarity(bcModel,Weighting.WEIGHTED);
            //                        //UserSimilarity similarity = new LogLikelihoodSimilarity(bcModel);
            //                        //UserSimilarity similarity = new PearsonCorrelationSimilarity(bcModel);
            //                        new TanimotoCoefficientSimilarity(dataModel)      
            //                  ),
            //                  0.5);
        }
    };

    RecommenderIRStatsEvaluator evaluator = new GenericRecommenderIRStatsEvaluator();
    IRStatistics stats = evaluator.evaluate(builder, null, model, null, 5, 1.0, 1.0);

    System.out.println(stats.getPrecision() + ", " + stats.getRecall() + ", " + stats.getF1Measure());

    //      assertEquals(0.75, stats.getPrecision(), EPSILON);
    //      assertEquals(0.75, stats.getRecall(), EPSILON);
    //      assertEquals(0.75, stats.getF1Measure(), EPSILON);
}

---

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

License:Apache License

@Test
public void evaluate5() {
    // RecommenderIRStatsEvaluator
    // ItemBasedRecommenderBuilder
    // TanimotoCoefficientSimilarity

    // IRStatistics stats =
    // evaluator.evaluate(builder, myModel, null, 3,
    // RecommenderIRStatusEvaluator.CHOOSE_THRESHOLD,
    // §1.0);

    try {/*from w  w w .  j  a v a 2s  .  c om*/
        DataModel model = createDataBooleanModel(input);

        RecommenderIRStatsEvaluator evaluator = new GenericRecommenderIRStatsEvaluator();
        IRStatistics evaluation = evaluator.evaluate(
                new ItemBasedRecommenderBuilder(TanimotoCoefficientSimilarity.class.getCanonicalName()),
                new BooleanDataModelBuilder(), model, null, 3,
                GenericRecommenderIRStatsEvaluator.CHOOSE_THRESHOLD, 0.9);
        LOG.info("result: " + evaluation);

        writetofile("ItemBasedRecommenderBuilder,TanimotoCoefficientSimilarity,RecommenderIRStatsEvaluator-F1,"
                + evaluation.getF1Measure() + "\n");
        // getFNMeasure
        // writetofile("ItemBasedRecommenderBuilder,EuclideanDistanceSimilarity,RecommenderIRStatsEvaluator-F1,"
        // + evaluation.getFNMeasure(n) + "\n");
        writetofile(
                "ItemBasedRecommenderBuilder,TanimotoCoefficientSimilarity,RecommenderIRStatsEvaluator-FallOut,"
                        + evaluation.getFallOut() + "\n");
        writetofile(
                "ItemBasedRecommenderBuilder,TanimotoCoefficientSimilarity,RecommenderIRStatsEvaluator-precision,"
                        + evaluation.getPrecision() + "\n");
        writetofile(
                "ItemBasedRecommenderBuilder,TanimotoCoefficientSimilarity,RecommenderIRStatsEvaluator-recall,"
                        + evaluation.getRecall() + "\n");

        evaluator = new GenericRecommenderIRStatsEvaluator();
        evaluation = evaluator.evaluate(
                new UserBasedRecommenderBuilder(TanimotoCoefficientSimilarity.class.getCanonicalName()),
                new BooleanDataModelBuilder(), model, null, 3, 3, 0.9);
        LOG.info("result: " + evaluation);

        writetofile("UserBasedRecommenderBuilder,TanimotoCoefficientSimilarity,RecommenderIRStatsEvaluator-F1,"
                + evaluation.getF1Measure() + "\n");
        // getFNMeasure
        // writetofile("ItemBasedRecommenderBuilder,EuclideanDistanceSimilarity,RecommenderIRStatsEvaluator-F1,"
        // + evaluation.getFNMeasure(n) + "\n");
        writetofile(
                "UserBasedRecommenderBuilder,TanimotoCoefficientSimilarity,RecommenderIRStatsEvaluator-FallOut,"
                        + evaluation.getFallOut() + "\n");
        writetofile(
                "UserBasedRecommenderBuilder,TanimotoCoefficientSimilarity,RecommenderIRStatsEvaluator-precision,"
                        + evaluation.getPrecision() + "\n");
        writetofile(
                "UserBasedRecommenderBuilder,TanimotoCoefficientSimilarity,RecommenderIRStatsEvaluator-recall,"
                        + evaluation.getRecall() + "\n");

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

}

---