org.grouplens.lenskit.predict.ordrec

Class OrdRecRatingPredictor

java.lang.Object

org.grouplens.lenskit.basic.AbstractRatingPredictor

org.grouplens.lenskit.predict.ordrec.OrdRecRatingPredictor

All Implemented Interfaces:

RatingPredictor

public class OrdRecRatingPredictor
extends AbstractRatingPredictor

Rating predictor using OrdRec to map scores to ratings. The model views user feedback as ordinal. The framework is based on a pointwise (rather than pairwise) ordinal approach, it can wrap existing CF methods, and upgrade them into being able to tackle ordinal feedback. The implementation is based on Koren's paper:

Since:

2.1

Field Summary

Fields

Modifier and Type	Field and Description
static TypedSymbol<mikera.vectorz.IVector>	RATING_PROBABILITY_CHANNEL

Constructor Summary

Constructors

Constructor and Description
OrdRecRatingPredictor(ItemScorer scorer, UserEventDAO dao, Quantizer q)
OrdRecRatingPredictor(ItemScorer scorer, UserEventDAO dao, Quantizer quantizer, double rate, double reg, int niters, boolean reportDist) Constructor of OrdRecRatingPrediciton

Method Summary

Modifier and Type	Method and Description
void	predict(long uid, MutableSparseVector predictions) Predict for items in a vector.

Methods inherited from class org.grouplens.lenskit.basic.AbstractRatingPredictor

predict, predict

Methods inherited from class java.lang.Object

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

Field Detail

RATING_PROBABILITY_CHANNEL

public static final TypedSymbol<mikera.vectorz.IVector> RATING_PROBABILITY_CHANNEL

Constructor Detail

OrdRecRatingPredictor

@Inject
public OrdRecRatingPredictor(ItemScorer scorer,
                                     UserEventDAO dao,
                                     Quantizer quantizer,
                                     @LearningRate
                                     double rate,
                                     @RegularizationTerm
                                     double reg,
                                     @IterationCount
                                     int niters,
                                     @ReportRatingDistribution
                                     boolean reportDist)

Constructor of OrdRecRatingPrediciton

Parameters:

scorer - The ItemScorer to create scores of each item

dao - The UserEventDAO

quantizer - The Quantizer used to quantizer the rating values

rate - The LearningRate

reg - The Regularization

OrdRecRatingPredictor

public OrdRecRatingPredictor(ItemScorer scorer,
                             UserEventDAO dao,
                             Quantizer q)

Method Detail

predict

public void predict(long uid,
                    @Nonnull
                    MutableSparseVector predictions)

Description copied from interface: RatingPredictor

Predict for items in a vector. The key domain of the provided vector is the items whose predictions are requested, and the predict method sets the values for each item to its predict (or unsets it, if no prediction can be provided). The previous values are discarded.

Parameters:

uid - The user ID.

predictions - The prediction output vector. Its key domain is the items to score.