es.upm.etsisi.cf4j.qualityMeasure.recommendation

Class Diversity

java.lang.Object

es.upm.etsisi.cf4j.qualityMeasure.QualityMeasure

es.upm.etsisi.cf4j.qualityMeasure.recommendation.Diversity

public class Diversity
extends QualityMeasure

This class the averaged diversity of the recomendations. Diversity value is computed as explained in "Hurley, N., & Zhang, M. (2011). Novelty and diversity in top-n recommendation--analysis and evaluation. ACM Transactions on Internet Technology (TOIT), 10(4), 1-30.". Lower values denotes more diverse recommendations.

Field Summary

Fields inherited from class es.upm.etsisi.cf4j.qualityMeasure.QualityMeasure

recommender

Constructor Summary

Constructors

Constructor and Description
Diversity(Recommender recommender, int numberOfRecommendations) Constructor of Diversity
Diversity(Recommender recommender, Map<String,Object> params) Constructor from a Map object with the quality measure parameters.

Method Summary

Modifier and Type	Method and Description
protected double	getScore(TestUser testUser, double[] predictions) Computes the quality measure score for a TestUser given the predictions for his/her test ratings

Methods inherited from class es.upm.etsisi.cf4j.qualityMeasure.QualityMeasure

getScore, getScore

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

Diversity

public Diversity(Recommender recommender,
                 Map<String,Object> params)

Constructor from a Map object with the quality measure parameters. Map object must contains the following keys:

• numberOfRecommendations: int value with the number of items to be recommended.

Parameters:

recommender - Recommender instance for which the Diversity are going to be computed

params - Quality measure's parameters

Diversity

public Diversity(Recommender recommender,
                 int numberOfRecommendations)

Constructor of Diversity

Parameters:

recommender - Recommender instance for which the precision are going to be computed

numberOfRecommendations - Number of recommendations. It must be greater than 1

Method Detail

getScore

protected double getScore(TestUser testUser,
                          double[] predictions)

Description copied from class: QualityMeasure

Computes the quality measure score for a TestUser given the predictions for his/her test ratings

Specified by:

getScore in class QualityMeasure

Parameters:

testUser - TestUser for which the quality measure score is computed

predictions - Prediction value for the test items rated by the test user

Returns:

Quality measure score