org.lenskit.basic

Class RescoringItemRecommender

java.lang.Object

org.lenskit.basic.RescoringItemRecommender

All Implemented Interfaces:

ItemRecommender

public class RescoringItemRecommender
extends java.lang.Object
implements ItemRecommender

Item recommender that wraps a base item recommender and replaces its scores with those produced by another item scorer. The order from the original recommender is preserved.

For performance reasons, the scorer is only invoked by recommendWithDetails(long, int, Set, Set), because scores are not returend in the other recommendation operations. The results produced by this recommender are of type RescoredResult; for any recommended item that the scorer cannot score, the result has no score.

See Also:

Results.rescore(Result, Result)

Constructor Summary

Constructors

Constructor and Description
RescoringItemRecommender(ItemRecommender rec, ItemScorer score) Create a new rescoring item recommender.

Method Summary

Modifier and Type	Method and Description
java.util.List<java.lang.Long>	recommend(long user) Recommend all possible items for a user using the default exclude set.
java.util.List<java.lang.Long>	recommend(long user, int n) Recommend up to n items for a user using the default exclude set.
java.util.List<java.lang.Long>	recommend(long user, int n, java.util.Set<java.lang.Long> candidates, java.util.Set<java.lang.Long> exclude) Produce a set of recommendations for the user.
ResultList	recommendWithDetails(long user, int n, java.util.Set<java.lang.Long> candidates, java.util.Set<java.lang.Long> exclude) Produce a set of recommendations for the user with additional details.

Methods inherited from class java.lang.Object

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

Constructor Detail

RescoringItemRecommender

@Inject
public RescoringItemRecommender(ItemRecommender rec,
                                        ItemScorer score)

Create a new rescoring item recommender.

Parameters:

rec - The recommender.

score - The item scorer.

Method Detail

recommend

public java.util.List<java.lang.Long> recommend(long user)

Description copied from interface: ItemRecommender

Recommend all possible items for a user using the default exclude set.

Specified by:

recommend in interface ItemRecommender

Parameters:

user - The user ID.

Returns:

The recommended items.

See Also:

ItemRecommender.recommend(long, int, Set, Set)

recommend

public java.util.List<java.lang.Long> recommend(long user,
                                                int n)

Description copied from interface: ItemRecommender

Recommend up to n items for a user using the default exclude set.

Specified by:

recommend in interface ItemRecommender

Parameters:

user - The user ID.

n - The number of recommendations to return. Negative values request as many recommendations as possible.

Returns:

The recommended items.

See Also:

ItemRecommender.recommend(long, int, Set, Set)

recommend

public java.util.List<java.lang.Long> recommend(long user,
                                                int n,
                                                @Nullable
                                                java.util.Set<java.lang.Long> candidates,
                                                @Nullable
                                                java.util.Set<java.lang.Long> exclude)

Description copied from interface: ItemRecommender

Produce a set of recommendations for the user. This is the most general recommendation method, allowing the recommendations to be constrained by both a candidate set \(\mathcal{C}\) and an exclude set \(\mathcal{E}\). The exclude set is applied to the candidate set, so the final effective candidate set is \(\mathcal{C} \backslash \mathcal{E}\).

The recommender is not guaranteed to return a full n recommendations. There are many reasons why it might return a shorter list, including lack of items, lack of coverage for items, or a predefined notion of a maximum recommendation list length. However, a negative value for n instructs the recommender to return as many as it can consistent with any limitations built in to its design and/or supporting algorithms.

Specified by:

recommend in interface ItemRecommender

Parameters:

user - The user’s ID

n - The number of ratings to return. If negative, the recommender will return as many recommendations as possible.

candidates - A set of candidate items which can be recommended. If null, all items are considered candidates.

exclude - A set of items to be excluded. If null, a default exclude set is used.

Returns:

A list of recommended items.

recommendWithDetails

public ResultList recommendWithDetails(long user,
                                       int n,
                                       @Nullable
                                       java.util.Set<java.lang.Long> candidates,
                                       @Nullable
                                       java.util.Set<java.lang.Long> exclude)

Description copied from interface: ItemRecommender

Produce a set of recommendations for the user with additional details. This method functions identically to ItemRecommender.recommend(long, int, Set, Set), except that it may produce more detailed results. Implementations may return subclasses of ResultList that provide access to additional details about each recommendation.

Specified by:

recommendWithDetails in interface ItemRecommender

Parameters:

user - The user’s ID

n - The number of ratings to return. If negative, then the recommender will return as many recommendations as possible.

candidates - A set of candidate items which can be recommended. If null, all items are considered candidates.

exclude - A set of items to be excluded. If null, a default exclude set is used.

Returns:

A list of recommended items. If the recommender cannot assign meaningful scores, the scores will be Double.NaN. For most scoring recommenders, the items will be ordered in decreasing order of score. This is not a hard requirement — e.g. set recommenders are allowed to be more flexible.