Class RandomStarRatingRecommender<TInstanceSource, TInstance, TUser, TItem, TDataRating, TFeatureSource, TFeatureValues>

Represents a star rating recommender system which generates predictions purely by random guessing.

Inheritance

Object

RandomStarRatingRecommender<TInstanceSource, TInstance, TUser, TItem, TDataRating, TFeatureSource, TFeatureValues>

Implements

IRecommender<TInstanceSource, TUser, TItem, Int32, IDictionary<Int32, Double>, TFeatureSource>

ILearner

Inherited Members

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

[SerializationVersion(1)]
public class RandomStarRatingRecommender<TInstanceSource, TInstance, TUser, TItem, TDataRating, TFeatureSource, TFeatureValues> : IRecommender<TInstanceSource, TUser, TItem, int, IDictionary<int, double>, TFeatureSource>, ILearner

Type Parameters

Name	Description
TInstanceSource	The type of a source of instances.
TInstance	The type of an instance.
TUser	The type of a user.
TItem	The type of an item.
TDataRating	The type of a rating in data.
TFeatureSource	The type of a feature source.
TFeatureValues	The type of the feature values.

Constructors

RandomStarRatingRecommender(IStarRatingRecommenderMapping<TInstanceSource, TInstance, TUser, TItem, TDataRating, TFeatureSource, TFeatureValues>)

Initializes a new instance of the RandomStarRatingRecommender<TInstanceSource, TInstance, TUser, TItem, TDataRating, TFeatureSource, TFeatureValues> class.

Declaration

public RandomStarRatingRecommender(IStarRatingRecommenderMapping<TInstanceSource, TInstance, TUser, TItem, TDataRating, TFeatureSource, TFeatureValues> mapping)

Parameters

Type	Name	Description
IStarRatingRecommenderMapping<TInstanceSource, TInstance, TUser, TItem, TDataRating, TFeatureSource, TFeatureValues>	mapping	The mapping used to access the data.

Exceptions

Type	Condition
ArgumentNullException	Thrown if one of the required parameters is null.

Properties

Capabilities

Gets the capabilities of the recommender.

Declaration

public IRecommenderCapabilities Capabilities { get; }

Property Value

Type	Description
IRecommenderCapabilities

ItemSubset

Gets or sets the subset of the items used for related item prediction and item recommendation.

Declaration

public IEnumerable<TItem> ItemSubset { get; set; }

Property Value

Type	Description
IEnumerable<TItem>

Settings

Gets or sets the settings (not supported).

Declaration

public ISettings Settings { get; set; }

Property Value

Type	Description
ISettings

UserSubset

Gets or sets the subset of the users used for related user prediction.

Declaration

public IEnumerable<TUser> UserSubset { get; set; }

Property Value

Type	Description
IEnumerable<TUser>

Methods

GetRelatedItems(TItem, Int32, TFeatureSource)

Returns a list of items related to item by randomly permuting the ItemSubset.

Declaration

public IEnumerable<TItem> GetRelatedItems(TItem item, int relatedItemCount, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
TItem	item	The item for which related items should be found.
Int32	relatedItemCount	Maximum number of related items to return.
TFeatureSource	featureSource	The source of features for the items.

Returns

Type	Description
IEnumerable<TItem>	The list of related items.

Remarks

Only items specified in ItemSubset will be returned.

GetRelatedItems(IEnumerable<TItem>, Int32, TFeatureSource)

Returns a list of related items to each item in items by randomly permuting the ItemSubset.

Declaration

public IDictionary<TItem, IEnumerable<TItem>> GetRelatedItems(IEnumerable<TItem> items, int relatedItemCount, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
IEnumerable<TItem>	items	The list of items for which related items should be found.
Int32	relatedItemCount	Maximum number of related items to return for every item.
TFeatureSource	featureSource	The source of features for the specified items.

Returns

Type	Description
IDictionary<TItem, IEnumerable<TItem>>	The list of related items for each item from items.

Remarks

Only items specified in ItemSubset will be returned.

GetRelatedUsers(TUser, Int32, TFeatureSource)

Returns a list of users related to user by randomly permuting the UserSubset.

Declaration

public IEnumerable<TUser> GetRelatedUsers(TUser user, int relatedUserCount, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
TUser	user	The user for which related users should be found.
Int32	relatedUserCount	Maximum number of related users to return.
TFeatureSource	featureSource	The source of features for the users.

Returns

Type	Description
IEnumerable<TUser>	The list of related users.

Remarks

Only users specified in UserSubset will be returned.

GetRelatedUsers(IEnumerable<TUser>, Int32, TFeatureSource)

Returns a list of related users to each user in users by randomly permuting the UserSubset.

Declaration

public IDictionary<TUser, IEnumerable<TUser>> GetRelatedUsers(IEnumerable<TUser> users, int relatedUserCount, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
IEnumerable<TUser>	users	The list of users for which related users should be found.
Int32	relatedUserCount	Maximum number of related users to return for every user.
TFeatureSource	featureSource	The source of features for the specified users.

Returns

Type	Description
IDictionary<TUser, IEnumerable<TUser>>	The list of related users for each user from users.

Remarks

Only users specified in UserSubset will be returned.

Predict(TInstanceSource, TFeatureSource)

Predicts ratings for the specified instances by random guessing.

Declaration

public IDictionary<TUser, IDictionary<TItem, int>> Predict(TInstanceSource instanceSource, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
TInstanceSource	instanceSource	The instances to predict ratings for.
TFeatureSource	featureSource	The source of features for the specified instances.

Returns

Type	Description
IDictionary<TUser, IDictionary<TItem, Int32>>	The predicted ratings.

Predict(TUser, TItem, TFeatureSource)

Predicts rating for a given user and item by random guessing.

Declaration

public int Predict(TUser user, TItem item, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
TUser	user	The user.
TItem	item	The item.
TFeatureSource	featureSource	The source of features for the specified user and item.

Returns

Type	Description
Int32	The predicted rating.

PredictDistribution(TInstanceSource, TFeatureSource)

Predicts rating distributions for the specified instances by returning the uniform distribution over the possible rating values.

Declaration

public IDictionary<TUser, IDictionary<TItem, IDictionary<int, double>>> PredictDistribution(TInstanceSource instanceSource, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
TInstanceSource	instanceSource	The instances to predict ratings for.
TFeatureSource	featureSource	The source of features for the specified instances.

Returns

Type	Description
IDictionary<TUser, IDictionary<TItem, IDictionary<Int32, Double>>>	The distributions of the predicted ratings.

PredictDistribution(TUser, TItem, TFeatureSource)

Predicts the distribution of a rating for a given user and item by returning a random distribution over ratings.

Declaration

public IDictionary<int, double> PredictDistribution(TUser user, TItem item, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
TUser	user	The user.
TItem	item	The item.
TFeatureSource	featureSource	The source of features for the given user and item.

Returns

Type	Description
IDictionary<Int32, Double>	The distribution of the rating.

Recommend(TUser, Int32, TFeatureSource)

Recommends items to a given user by randomly permuting the ItemSubset.

Declaration

public IEnumerable<TItem> Recommend(TUser user, int recommendationCount, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
TUser	user	The user to recommend items to.
Int32	recommendationCount	Maximum number of items to recommend.
TFeatureSource	featureSource	The source of features for the specified user.

Returns

Type	Description
IEnumerable<TItem>	The list of recommended items.

Remarks

Only items specified in ItemSubset can be recommended.

Recommend(IEnumerable<TUser>, Int32, TFeatureSource)

Recommends items to a given list of users by randomly permuting the ItemSubset.

Declaration

public IDictionary<TUser, IEnumerable<TItem>> Recommend(IEnumerable<TUser> users, int recommendationCount, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
IEnumerable<TUser>	users	The list of users to recommend items to.
Int32	recommendationCount	Maximum number of items to recommend to a single user.
TFeatureSource	featureSource	The source of features for the specified users.

Returns

Type	Description
IDictionary<TUser, IEnumerable<TItem>>	The list of recommended items for every user from users.

Remarks

Only items specified in ItemSubset can be recommended.

RecommendDistribution(TUser, Int32, TFeatureSource)

Recommend items with their rating distributions to a specified user.

Declaration

public IEnumerable<Tuple<TItem, IDictionary<int, double>>> RecommendDistribution(TUser user, int recommendationCount, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
TUser	user	The user to recommend items to.
Int32	recommendationCount	Maximum number of items to recommend.
TFeatureSource	featureSource	The source of features for the specified user.

Returns

Type	Description
IEnumerable<Tuple<TItem, IDictionary<Int32, Double>>>	The list of recommended items and their rating distributions.

Remarks

Only items specified in ItemSubset can be recommended.

RecommendDistribution(IEnumerable<TUser>, Int32, TFeatureSource)

Recommends items with their rating distributions to a specified list of users.

Declaration

public IDictionary<TUser, IEnumerable<Tuple<TItem, IDictionary<int, double>>>> RecommendDistribution(IEnumerable<TUser> users, int recommendationCount, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
IEnumerable<TUser>	users	The list of users to recommend items to.
Int32	recommendationCount	Maximum number of items to recommend to a single user.
TFeatureSource	featureSource	The source of features for the specified users.

Returns

Type	Description
IDictionary<TUser, IEnumerable<Tuple<TItem, IDictionary<Int32, Double>>>>	The list of recommended items and their rating distributions for every user from users.

Remarks

Only items specified in ItemSubset can be recommended.

Train(TInstanceSource, TFeatureSource)

Trains the recommender on the specified instances. For the random recommender it results in just retrieving the rating info, as well as the list of items and users from the training set.

Declaration

public void Train(TInstanceSource instanceSource, TFeatureSource featureSource = null)

Parameters

Type	Name	Description
TInstanceSource	instanceSource	The source of instances to train on.
TFeatureSource	featureSource	The source of features for the specified instances.

Explicit Interface Implementations

ILearner.Capabilities

Gets the capabilities of the learner. These are any properties of the learner that are not captured by the type signature of the most specific learner interface below.

Declaration

ICapabilities ILearner.Capabilities { get; }

Returns

Type	Description
ICapabilities

Implements

IRecommender<TInstanceSource, TUser, TItem, TRating, TRatingDistribution, TFeatureSource>

ILearner