org.lenskit.mf.svd

Class BiasedMFItemScorer

java.lang.Object

org.lenskit.basic.AbstractItemScorer

org.lenskit.mf.svd.BiasedMFItemScorer

All Implemented Interfaces:

ItemScorer

Direct Known Subclasses:

FunkSVDItemScorer

public class BiasedMFItemScorer
extends AbstractItemScorer

Item scorer using biased matrix factorization. This implements SVD-style item scorers.

Since:

2.1

Constructor Summary

Constructors

Constructor and Description
BiasedMFItemScorer(MFModel mod, BiasedMFKernel kern, ItemScorer bl) Create a new biased MF item scorer.

Method Summary

Modifier and Type	Method and Description
MFModel	getModel()
protected org.apache.commons.math3.linear.RealVector	getUserPreferenceVector(long user) Get a user’s preference vector.
ResultMap	scoreWithDetails(long user, Collection<Long> items) Score a collection of items and potentially return more details on the scores.

Methods inherited from class org.lenskit.basic.AbstractItemScorer

score, score

Methods inherited from class java.lang.Object

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

Constructor Detail

BiasedMFItemScorer

@Inject
public BiasedMFItemScorer(MFModel mod,
                                   BiasedMFKernel kern,
                                   @BaselineScorer
                                   ItemScorer bl)

Create a new biased MF item scorer.

Parameters:

mod - The model (factorized matrix)

kern - The kernel function to compute scores.

bl - The baseline scorer (used to compute biases).

Method Detail

getUserPreferenceVector

@Nullable
protected org.apache.commons.math3.linear.RealVector getUserPreferenceVector(long user)

Get a user’s preference vector.

Parameters:

user - The user ID.

Returns:

The user’s preference vector, or null if no preferences are available for the user.

getModel

public MFModel getModel()

scoreWithDetails

@Nonnull
public ResultMap scoreWithDetails(long user,
                                            @Nonnull
                                            Collection<Long> items)

Description copied from interface: ItemScorer

Score a collection of items and potentially return more details on the scores.

Parameters:

user - The user ID for whom to generate scores.

items - The item to score.

Returns:

The scores for the items. This result set may not contain all requested items. Implementations that support additional details will return a subclass of ResultMap that provides access to those details.