org.grouplens.lenskit.eval.metrics.topn

Class PrecisionRecallTopNMetric

java.lang.Object

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

org.grouplens.lenskit.eval.metrics.topn.PrecisionRecallTopNMetric

All Implemented Interfaces:

Closeable, AutoCloseable, Metric<PrecisionRecallTopNMetric.Context>

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

A metric to compute the precision and recall of a recommender given a set of candidate items to recommend from and a set of desired items. The aggregate results are means of the user results. This can be used to compute metrics like fallout (probability that a recommendation is bad) by configuring bad items as the test item set.

Nested Class Summary

Nested Classes

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

Constructor Summary

Constructors

Constructor and Description
PrecisionRecallTopNMetric(String pre, String sfx, int listSize, ItemSelector candidates, ItemSelector exclude, ItemSelector goodItems) Construct a new recall and precision top n metric

Method Summary

Modifier and Type	Method and Description
PrecisionRecallTopNMetric.Context	createContext(Attributed algo, TTDataSet ds, Recommender rec) Create the context for an experimental condition (algorithm/data set pair).
PrecisionRecallTopNMetric.Result	doMeasureUser(TestUser user, PrecisionRecallTopNMetric.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 PrecisionRecallTopNMetric.Result	getTypedResults(PrecisionRecallTopNMetric.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

PrecisionRecallTopNMetric

public PrecisionRecallTopNMetric(String pre,
                                 String sfx,
                                 int listSize,
                                 ItemSelector candidates,
                                 ItemSelector exclude,
                                 ItemSelector goodItems)

Construct a new recall and precision top n metric

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.

listSize - The number of recommendations to fetch.

candidates - The candidate selector, provides a list of items which can be recommended

exclude - The exclude selector, provides a list of items which must not be recommended (These items are removed from the candidate items to form the final candidate set)

goodItems - The list of items to consider "true positives", all other items will be treated as "false positives".

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<PrecisionRecallTopNMetric.Context,PrecisionRecallTopNMetric.Result,PrecisionRecallTopNMetric.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<PrecisionRecallTopNMetric.Context,PrecisionRecallTopNMetric.Result,PrecisionRecallTopNMetric.Result>

Returns:

The suffix to apply to column names.

createContext

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

Description copied from class: AbstractMetric

Measure a user with typed results.

Specified by:

doMeasureUser in class AbstractMetric<PrecisionRecallTopNMetric.Context,PrecisionRecallTopNMetric.Result,PrecisionRecallTopNMetric.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 PrecisionRecallTopNMetric.Result getTypedResults(PrecisionRecallTopNMetric.Context context)

Description copied from class: AbstractMetric

Get the typed results from an accumulator.

Specified by:

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

Parameters:

context - The context.

Returns:

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