Class StarRatingRecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating>

Evaluates a recommender system which predicts star ratings.

Inheritance

Object

RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, Int32, IDictionary<Int32, Double>>

StarRatingRecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating>

Inherited Members

RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, Int32, IDictionary<Int32, Double>>.RecommendRatedItems<TFeatureSource>(IRecommender<TInstanceSource, TUser, TItem, Int32, IDictionary<Int32, Double>, TFeatureSource>, TInstanceSource, Int32, Int32, TFeatureSource)

RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, Int32, IDictionary<Int32, Double>>.FindRelatedUsersWhoRatedSameItems<TFeatureSource>(IRecommender<TInstanceSource, TUser, TItem, Int32, IDictionary<Int32, Double>, TFeatureSource>, TInstanceSource, Int32, Int32, Int32, TFeatureSource)

RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, Int32, IDictionary<Int32, Double>>.FindRelatedItemsRatedBySameUsers<TFeatureSource>(IRecommender<TInstanceSource, TUser, TItem, Int32, IDictionary<Int32, Double>, TFeatureSource>, TInstanceSource, Int32, Int32, Int32, TFeatureSource)

RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, Int32, IDictionary<Int32, Double>>.RatingPredictionMetric(TInstanceSource, IDictionary<TUser, IDictionary<TItem, Int32>>, Func<TGroundTruthRating, Int32, Double>, RecommenderMetricAggregationMethod)

RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, Int32, IDictionary<Int32, Double>>.RatingPredictionMetric(TInstanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>, Func<TGroundTruthRating, IDictionary<Int32, Double>, Double>, RecommenderMetricAggregationMethod)

RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, Int32, IDictionary<Int32, Double>>.ItemRecommendationMetric(TInstanceSource, IDictionary<TUser, IEnumerable<TItem>>, Func<IEnumerable<Double>, Double>, Func<TGroundTruthRating, Double>)

RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, Int32, IDictionary<Int32, Double>>.ItemRecommendationMetric(TInstanceSource, IDictionary<TUser, IEnumerable<TItem>>, Func<IEnumerable<Double>, IEnumerable<Double>, Double>, Func<TGroundTruthRating, Double>)

RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, Int32, IDictionary<Int32, Double>>.RelatedUsersMetric(TInstanceSource, IDictionary<TUser, IEnumerable<TUser>>, Int32, Func<IEnumerable<Double>, IEnumerable<Double>, Double>, Func<Vector, Vector, Double>, Func<TGroundTruthRating, Double>)

RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, Int32, IDictionary<Int32, Double>>.RelatedUsersMetric(TInstanceSource, IDictionary<TUser, IEnumerable<TUser>>, Int32, Func<IEnumerable<Double>, Double>, Func<Vector, Vector, Double>, Func<TGroundTruthRating, Double>)

RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, Int32, IDictionary<Int32, Double>>.RelatedItemsMetric(TInstanceSource, IDictionary<TItem, IEnumerable<TItem>>, Int32, Func<IEnumerable<Double>, IEnumerable<Double>, Double>, Func<Vector, Vector, Double>, Func<TGroundTruthRating, Double>)

RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, Int32, IDictionary<Int32, Double>>.RelatedItemsMetric(TInstanceSource, IDictionary<TItem, IEnumerable<TItem>>, Int32, Func<IEnumerable<Double>, Double>, Func<Vector, Vector, Double>, Func<TGroundTruthRating, Double>)

Object.Equals(Object)

Object.Equals(Object, Object)

Object.GetHashCode()

Object.GetType()

Object.MemberwiseClone()

Object.ReferenceEquals(Object, Object)

Object.ToString()

Namespace: Microsoft.ML.Probabilistic.Learners

Assembly: Microsoft.ML.Probabilistic.Learners.Recommender.dll

Syntax

public class StarRatingRecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating> : RecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating, int, IDictionary<int, double>>

Type Parameters

Name	Description
TInstanceSource	The type of a source of instances.
TUser	The type of a user.
TItem	The type of an item.
TGroundTruthRating	The type of a rating in a test dataset.

Constructors

StarRatingRecommenderEvaluator(IStarRatingRecommenderEvaluatorMapping<TInstanceSource, TUser, TItem, TGroundTruthRating>)

Initializes a new instance of the StarRatingRecommenderEvaluator<TInstanceSource, TUser, TItem, TGroundTruthRating> class.

Declaration

public StarRatingRecommenderEvaluator(IStarRatingRecommenderEvaluatorMapping<TInstanceSource, TUser, TItem, TGroundTruthRating> mapping)

Parameters

Type	Name	Description
IStarRatingRecommenderEvaluatorMapping<TInstanceSource, TUser, TItem, TGroundTruthRating>	mapping	The mapping used for accessing data.

Methods

ConfusionMatrix(TInstanceSource, IDictionary<TUser, IDictionary<TItem, Int32>>, RecommenderMetricAggregationMethod)

Computes a confusion matrix.

Declaration

public RatingMatrix ConfusionMatrix(TInstanceSource instanceSource, IDictionary<TUser, IDictionary<TItem, int>> predictions, RecommenderMetricAggregationMethod aggregationMethod = RecommenderMetricAggregationMethod.Default)

Parameters

Type	Name	Description
TInstanceSource	instanceSource	The instance source providing the ground truth.
IDictionary<TUser, IDictionary<TItem, Int32>>	predictions	A sparse users-by-items matrix of predicted rating distributions.
RecommenderMetricAggregationMethod	aggregationMethod	A method specifying how metrics are aggregated over the whole dataset.

Returns

Type	Description
RatingMatrix	The computed confusion matrix.

ExpectedConfusionMatrix(TInstanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>, RecommenderMetricAggregationMethod)

Computes the expected confusion matrix.

Declaration

public RatingMatrix ExpectedConfusionMatrix(TInstanceSource instanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<int, double>>> predictions, RecommenderMetricAggregationMethod aggregationMethod = RecommenderMetricAggregationMethod.Default)

Parameters

Type	Name	Description
TInstanceSource	instanceSource	The instance source providing the ground truth.
IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>	predictions	A sparse users-by-items matrix of predicted rating distributions.
RecommenderMetricAggregationMethod	aggregationMethod	A method specifying how metrics are aggregated over all instances.

Returns

Type	Description
RatingMatrix	The computed expected confusion matrix.

ExpectedWeightedConfusion(TInstanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>, RatingMatrix, RecommenderMetricAggregationMethod)

Computes the expected component-wise product of a confusion matrix and a loss matrix.

Declaration

public double ExpectedWeightedConfusion(TInstanceSource instanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<int, double>>> predictions, RatingMatrix lossMatrix, RecommenderMetricAggregationMethod aggregationMethod = RecommenderMetricAggregationMethod.Default)

Parameters

Type	Name	Description
TInstanceSource	instanceSource	The instance source providing the ground truth (used to compute the expected confusion matrix).
IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>	predictions	A sparse users-by-items matrix of predicted rating distributions (used to compute the expected confusion matrix).
RatingMatrix	lossMatrix	The loss matrix.
RecommenderMetricAggregationMethod	aggregationMethod	A method specifying how metrics are aggregated over all instances (used to compute the expected confusion matrix).

Returns

Type	Description
Double	The computed expected weighted confusion.

ModelDomainRatingPredictionMetric(TInstanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>, Func<Int32, IDictionary<Int32, Double>, Double>, RecommenderMetricAggregationMethod)

Computes the average of a given rating prediction metric using ground truth in model domain by iterating over predictions and using the aggregation method given in aggregationMethod.

Declaration

public double ModelDomainRatingPredictionMetric(TInstanceSource instanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<int, double>>> predictions, Func<int, IDictionary<int, double>, double> metric, RecommenderMetricAggregationMethod aggregationMethod = RecommenderMetricAggregationMethod.Default)

Parameters

Type	Name	Description
TInstanceSource	instanceSource	The instance source providing the ground truth.
IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>	predictions	A sparse users-by-items matrix of predicted rating distributions.
Func<Int32, IDictionary<Int32, Double>, Double>	metric	The rating prediction metric using ground truth in model domain.
RecommenderMetricAggregationMethod	aggregationMethod	A method specifying how metrics are aggregated over all instances.

Returns

Type	Description
Double	The computed average of the given rating prediction metric.

ModelDomainRatingPredictionMetric(TInstanceSource, IDictionary<TUser, IDictionary<TItem, Int32>>, Func<Int32, Int32, Double>, RecommenderMetricAggregationMethod)

Computes the average of a given rating prediction metric using ground truth in model domain by iterating over predictions and using the aggregation method given in aggregationMethod.

Declaration

public double ModelDomainRatingPredictionMetric(TInstanceSource instanceSource, IDictionary<TUser, IDictionary<TItem, int>> predictions, Func<int, int, double> metric, RecommenderMetricAggregationMethod aggregationMethod = RecommenderMetricAggregationMethod.Default)

Parameters

Type	Name	Description
TInstanceSource	instanceSource	The instance source providing the ground truth.
IDictionary<TUser, IDictionary<TItem, Int32>>	predictions	A sparse users-by-items matrix of predicted rating distributions.
Func<Int32, Int32, Double>	metric	The rating prediction metric using ground truth in model domain.
RecommenderMetricAggregationMethod	aggregationMethod	A method specifying how metrics are aggregated over all instances.

Returns

Type	Description
Double	The computed average of the given rating prediction metric.

ModelDomainRatingPredictionMetricExpectation(TInstanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>, Func<Int32, Int32, Double>, RecommenderMetricAggregationMethod)

Computes the average of a given rating prediction metric using ground truth in model domain by iterating over predictions and using the aggregation method given in aggregationMethod.

Declaration

public double ModelDomainRatingPredictionMetricExpectation(TInstanceSource instanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<int, double>>> predictions, Func<int, int, double> metric, RecommenderMetricAggregationMethod aggregationMethod = RecommenderMetricAggregationMethod.Default)

Parameters

Type	Name	Description
TInstanceSource	instanceSource	The instance source providing the ground truth.
IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>	predictions	A sparse users-by-items matrix of predicted rating distributions.
Func<Int32, Int32, Double>	metric	The rating prediction metric using ground truth in model domain.
RecommenderMetricAggregationMethod	aggregationMethod	A method specifying how metrics are aggregated over all instances.

Returns

Type	Description
Double	The computed average of the given rating prediction metric.

ProbabilityCalibrationError(TInstanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>, Int32, Int32)

Computes the probability calibration error for a given rating.

Declaration

public double ProbabilityCalibrationError(TInstanceSource instanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<int, double>>> predictions, int rating, int bins)

Parameters

Type	Name	Description
TInstanceSource	instanceSource	The instance source providing the ground truth.
IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>	predictions	A sparse users-by-items matrix of predicted rating distributions.
Int32	rating	The rating value to generate the calibration plot for.
Int32	bins	The number of bins to use.

Returns

Type	Description
Double	The computed probability calibration error.

ProbabilityCalibrationPlot(TInstanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>, Int32, Int32)

Computes the probability calibration plot for a particular rating value.

Declaration

public double[] ProbabilityCalibrationPlot(TInstanceSource instanceSource, IDictionary<TUser, IDictionary<TItem, IDictionary<int, double>>> predictions, int rating, int bins)

Parameters

Type	Name	Description
TInstanceSource	instanceSource	The instance source providing the ground truth.
IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>	predictions	A sparse users-by-items matrix of predicted rating distributions.
Int32	rating	The rating value to generate the calibration plot for.
Int32	bins	The number of bins to use.

Returns

Type	Description
Double[]	The computed probability calibration plot.

WeightedConfusion(TInstanceSource, IDictionary<TUser, IDictionary<TItem, Int32>>, RatingMatrix, RecommenderMetricAggregationMethod)

Computes the component-wise product of a confusion matrix and a loss matrix.

Declaration

public double WeightedConfusion(TInstanceSource instanceSource, IDictionary<TUser, IDictionary<TItem, int>> predictions, RatingMatrix lossMatrix, RecommenderMetricAggregationMethod aggregationMethod = RecommenderMetricAggregationMethod.Default)

Parameters

Type	Name	Description
TInstanceSource	instanceSource	The instance source providing the ground truth (used to compute the confusion matrix).
IDictionary<TUser, IDictionary<TItem, Int32>>	predictions	A sparse users-by-items matrix of predicted rating distributions (used to compute the confusion matrix).
RatingMatrix	lossMatrix	The loss matrix.
RecommenderMetricAggregationMethod	aggregationMethod	A method specifying how metrics are aggregated over all instances (used to compute the confusion matrix).

Returns

Type	Description
Double	The computed weighted confusion.