org.grouplens.lenskit.eval.metrics.topn

Class IndependentRecallTopNMetric

java.lang.Object

org.grouplens.lenskit.eval.metrics.AbstractMetric<IndependentRecallTopNMetric.Context,IndependentRecallTopNMetric.Result,IndependentRecallTopNMetric.Result>

org.grouplens.lenskit.eval.metrics.topn.IndependentRecallTopNMetric

All Implemented Interfaces:

Closeable, AutoCloseable, Metric<IndependentRecallTopNMetric.Context>

public class IndependentRecallTopNMetric
extends AbstractMetric<IndependentRecallTopNMetric.Context,IndependentRecallTopNMetric.Result,IndependentRecallTopNMetric.Result>

An alternate methodology from computing precision/recall scores. This is computed for each true positive item selected by the testItems selector. For each item a set of other items is chosen by the candidate set. A topN recommendation is then made over the candidate items (plus the test item). The item is considered a hit if the true positive item is in the recommendations. The hit rate over the set of testItems is reported. Under this methodology percussion and recall are roughly equivalent, so only recall (hit rate) is returned.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	IndependentRecallTopNMetric.Builder
class	IndependentRecallTopNMetric.Context
static class	IndependentRecallTopNMetric.Result

Constructor Summary

Constructors

Constructor and Description
IndependentRecallTopNMetric(String pre, String sfx, ItemSelector queryItems, ItemSelector candidates, int listSize, ItemSelector exclude)

Method Summary

Modifier and Type	Method and Description
IndependentRecallTopNMetric.Context	createContext(Attributed algo, TTDataSet ds, Recommender rec) Create the context for an experimental condition (algorithm/data set pair).
IndependentRecallTopNMetric.Result	doMeasureUser(TestUser user, IndependentRecallTopNMetric.Context context) Measure a user with typed results.
protected String	getPrefix() A prefix to be applied to column names.
protected String	getSuffix() A suffix to be applied to column names.
protected IndependentRecallTopNMetric.Result	getTypedResults(IndependentRecallTopNMetric.Context context) Get the typed results from an accumulator.

Methods inherited from class org.grouplens.lenskit.eval.metrics.AbstractMetric

close, getColumnLabels, getResults, getUserColumnLabels, measureUser

Methods inherited from class java.lang.Object

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

Constructor Detail

IndependentRecallTopNMetric

public IndependentRecallTopNMetric(String pre,
                                   String sfx,
                                   ItemSelector queryItems,
                                   ItemSelector candidates,
                                   int listSize,
                                   ItemSelector exclude)

Parameters:

pre - the prefix label for this evaluation, or null for no prefix.

sfx - the suffix label for this evaluation, or null for no suffix.

queryItems - the "true positive" items that we compute the hit rate over

candidates - items to add to the recommendation, should be a random selection

listSize - The size of the recommendation list to evaluate

exclude - Items which should not be included in the recommendations. Should not include test set.

Method Detail

getPrefix

protected String getPrefix()

Description copied from class: AbstractMetric

A prefix to be applied to column names. If non-null, this prefix will be applied and separated with a period.

Overrides:

getPrefix in class AbstractMetric<IndependentRecallTopNMetric.Context,IndependentRecallTopNMetric.Result,IndependentRecallTopNMetric.Result>

Returns:

The prefix to apply to column names.

getSuffix

protected String getSuffix()

Description copied from class: AbstractMetric

A suffix to be applied to column names. If non-null, this suffix will be applied and separated with a period.

Overrides:

getSuffix in class AbstractMetric<IndependentRecallTopNMetric.Context,IndependentRecallTopNMetric.Result,IndependentRecallTopNMetric.Result>

Returns:

The suffix to apply to column names.

createContext

public IndependentRecallTopNMetric.Context createContext(Attributed algo,
                                                         TTDataSet ds,
                                                         Recommender rec)

Description copied from interface: Metric

Create the context for an experimental condition (algorithm/data set pair).

Parameters:

algo - The algorithm.

ds - The data set.

rec - The LensKit recommender, if applicable. This can be null for an external algorithm that does not provide a LensKit recommender.

Returns:

The accumulator. This will be passed to Metric.measureUser(TestUser, Object). If the metric does not accumulate any results, this method can return null.

doMeasureUser

public IndependentRecallTopNMetric.Result doMeasureUser(TestUser user,
                                                        IndependentRecallTopNMetric.Context context)

Description copied from class: AbstractMetric

Measure a user with typed results.

Specified by:

doMeasureUser in class AbstractMetric<IndependentRecallTopNMetric.Context,IndependentRecallTopNMetric.Result,IndependentRecallTopNMetric.Result>

Parameters:

user - The user to measure.

context - The context.

Returns:

The results of measuring the user, or null to emit NAs for the user.

getTypedResults

protected IndependentRecallTopNMetric.Result getTypedResults(IndependentRecallTopNMetric.Context context)

Description copied from class: AbstractMetric

Get the typed results from an accumulator.

Specified by:

getTypedResults in class AbstractMetric<IndependentRecallTopNMetric.Context,IndependentRecallTopNMetric.Result,IndependentRecallTopNMetric.Result>

Parameters:

context - The context.

Returns:

The results accumulated for this experiment, or null to emit NAs.