Time-aware bias model. More...

Inheritance diagram for TimeAwareBaseline:

Public Member Functions
virtual bool	CanPredict (int user_id, int item_id)
	Check whether a useful prediction (i.e. not using a fallback/default answer) can be made for a given user-item combination.
Object	Clone ()
	create a shallow copy of the object
virtual float	ComputeObjective ()
	Compute the current optimization objective (usually loss plus regularization term) of the model.
virtual void	Iterate ()
	Run one iteration (= pass over the training data)
virtual void	LoadModel (string file)
	Get the model parameters from a file.
override float	Predict (int user_id, int item_id)
	Predict rating or score for a given user-item combination.
override float	Predict (int user_id, int item_id, DateTime time)
	predict rating at a certain point in time
virtual void	SaveModel (string file)
	Save the model parameters to a file.
	TimeAwareBaseline ()
	default constructor
override string	ToString ()
	Return a string representation of the recommender.
override void	Train ()
	Learn the model parameters of the recommender from the training data.
Protected Member Functions
virtual void	InitModel ()
	Initialize the model parameters.
virtual float	Predict (int user_id, int item_id, int day, int bin)
	Predict the specified user_id, item_id, day and bin.
int	RelativeDay (DateTime datetime)
	Given a DateTime object, return the day relative to the first rating day in the dataset.
virtual void	UpdateParameters (int u, int i, int day, int bin, float err)
	Single stochastic gradient descent step: update the parameter values for one user and one item.
Protected Attributes
int	latest_relative_day
	last day in the training data, counting from the first day
float	max_rating
	Maximum rating value.
float	min_rating
	Minimum rating value.
IRatings	ratings
	rating data
ITimedRatings	timed_ratings
	rating data, including time information
Properties
float	AlphaLearnRate `[get, set]`
	learn rate for the user-wise alphas
float	Beta `[get, set]`
	beta parameter for modeling the drift in the user bias
int	BinSize `[get, set]`
	bin size in days for modeling the time-dependent item bias
float	ItemBiasByTimeBinLearnRate `[get, set]`
	learn rate for the bin-wise item bias
float	ItemBiasLearnRate `[get, set]`
	learn rate for the item bias
int	MaxItemID `[get, set]`
	Maximum item ID.
virtual float	MaxRating `[get, set]`
	Maximum rating value.
int	MaxUserID `[get, set]`
	Maximum user ID.
virtual float	MinRating `[get, set]`
	Minimum rating value.
uint	NumIter `[get, set]`
	number of iterations over the dataset to perform
override IRatings	Ratings `[get, set]`
	The rating data.
float	RegAlpha `[get, set]`
	regularization for the user-wise alphas
float	RegI `[get, set]`
	regularization for the item bias
float	RegItemBiasByTimeBin `[get, set]`
	regularization for the bin-wise item bias
float	RegU `[get, set]`
	regularization for the user bias
float	RegUserBiasByDay `[get, set]`
	regularization for the day-wise user bias
float	RegUserScaling `[get, set]`
	regularization for the user scaling factor
float	RegUserScalingByDay `[get, set]`
	regularization for the day-wise user scaling factor
virtual ITimedRatings	TimedRatings `[get, set]`
	the rating data, including time information
float	UserBiasByDayLearnRate `[get, set]`
	learn rate for the day-wise user bias
float	UserBiasLearnRate `[get, set]`
	learn rate for the user bias
float	UserScalingByDayLearnRate `[get, set]`
	learn rate for the day-wise user scaling factor
float	UserScalingLearnRate `[get, set]`
	learn rate for the user-wise scaling factor

Detailed Description

Time-aware bias model.

Model described in equation (10) of BellKor Grand Prize documentation for the Netflix Prize (see below). The optimization problem is described in equation (12).

The default hyper-parameter values are set to the ones shown in the report. For datasets other than Netflix, you may want to find better parameters.

Literature:

Yehuda Koren: The BellKor Solution to the Netflix Grand Prize

This recommender does currently NOT support incremental updates.

Constructor & Destructor Documentation

TimeAwareBaseline ( ) [inline]

default constructor

Member Function Documentation

virtual bool CanPredict	(	int	user_id,
		int	item_id
	)		`[inline, virtual, inherited]`

Check whether a useful prediction (i.e. not using a fallback/default answer) can be made for a given user-item combination.

It is up to the recommender implementor to decide when a prediction is useful, and to document it accordingly.

Parameters:

user_id	the user ID
item_id	the item ID

Returns:: true if a useful prediction can be made, false otherwise

Implements IRecommender.

Reimplemented in BiPolarSlopeOne, Constant, SlopeOne, GlobalAverage, UserAverage, ItemAverage, and Random.

Object Clone ( ) [inline, inherited]

create a shallow copy of the object

virtual float ComputeObjective ( ) [inline, virtual]

Compute the current optimization objective (usually loss plus regularization term) of the model.

Returns:: the current objective; -1 if not implemented

Implements IIterativeModel.

Reimplemented in TimeAwareBaselineWithFrequencies.

virtual void InitModel ( ) [inline, protected, virtual]

Initialize the model parameters.

Reimplemented in TimeAwareBaselineWithFrequencies.

virtual void Iterate ( ) [inline, virtual]

Run one iteration (= pass over the training data)

Implements IIterativeModel.

virtual void LoadModel ( string filename ) [inline, virtual, inherited]

Get the model parameters from a file.

Parameters:

filename the name of the file to read from

Implements IRecommender.

Reimplemented in MatrixFactorization, BiasedMatrixFactorization, CoClustering, SVDPlusPlus, FactorWiseMatrixFactorization, UserItemBaseline, SigmoidCombinedAsymmetricFactorModel, SigmoidSVDPlusPlus, BiPolarSlopeOne, SigmoidItemAsymmetricFactorModel, SigmoidUserAsymmetricFactorModel, ItemKNN, NaiveBayes, SlopeOne, EntityAverage, KNN, GlobalAverage, Constant, and Random.

override float Predict	(	int	user_id,
		int	item_id
	)		`[inline, virtual]`

Predict rating or score for a given user-item combination.

Parameters:

user_id	the user ID
item_id	the item ID

Returns:: the predicted score/rating for the given user-item combination

Implements RatingPredictor.

virtual float Predict	(	int	user_id,
		int	item_id,
		int	day,
		int	bin
	)		`[inline, protected, virtual]`

Predict the specified user_id, item_id, day and bin.

Assumes user and item IDs are valid.

Parameters:

user_id	the user ID
item_id	the item ID
day	the day of the rating
bin	the day bin of the rating

Reimplemented in TimeAwareBaselineWithFrequencies.

override float Predict	(	int	user_id,
		int	item_id,
		DateTime	time
	)		`[inline, virtual]`

predict rating at a certain point in time

Parameters:

user_id	the user ID
item_id	the item ID
time	the time of the rating event

Implements TimeAwareRatingPredictor.

Reimplemented in TimeAwareBaselineWithFrequencies.

int RelativeDay ( DateTime datetime ) [inline, protected]

Given a DateTime object, return the day relative to the first rating day in the dataset.

Returns:: the day relative to the first rating day in the dataset

Parameters:

datetime the date/time of the rating event

virtual void SaveModel ( string filename ) [inline, virtual, inherited]

Save the model parameters to a file.

Parameters:

filename the name of the file to write to

Implements IRecommender.

Reimplemented in MatrixFactorization, BiasedMatrixFactorization, CoClustering, SVDPlusPlus, FactorWiseMatrixFactorization, UserItemBaseline, BiPolarSlopeOne, SigmoidCombinedAsymmetricFactorModel, NaiveBayes, SigmoidItemAsymmetricFactorModel, SigmoidUserAsymmetricFactorModel, SlopeOne, EntityAverage, KNN, GlobalAverage, Constant, and Random.

override string ToString ( ) [inline]

Return a string representation of the recommender.

The ToString() method of recommenders should list the class name and all hyperparameters, separated by space characters.

Reimplemented from RatingPredictor.

Reimplemented in TimeAwareBaselineWithFrequencies.

virtual void UpdateParameters	(	int	u,
		int	i,
		int	day,
		int	bin,
		float	err
	)		`[inline, protected, virtual]`

Single stochastic gradient descent step: update the parameter values for one user and one item.

Parameters:

u	the user ID
i	the item ID
day	the day of the rating
bin	the day bin of the rating
err	the current error made for this rating

Reimplemented in TimeAwareBaselineWithFrequencies.

Member Data Documentation

int latest_relative_day [protected]

last day in the training data, counting from the first day

float max_rating [protected, inherited]

Maximum rating value.

float min_rating [protected, inherited]

Minimum rating value.

IRatings ratings [protected, inherited]

rating data

ITimedRatings timed_ratings [protected, inherited]

rating data, including time information

Property Documentation

float AlphaLearnRate [get, set]

learn rate for the user-wise alphas

float Beta [get, set]

beta parameter for modeling the drift in the user bias

int BinSize [get, set]

bin size in days for modeling the time-dependent item bias

float ItemBiasByTimeBinLearnRate [get, set]

learn rate for the bin-wise item bias

float ItemBiasLearnRate [get, set]

learn rate for the item bias

int MaxItemID [get, set, inherited]

Maximum item ID.

virtual float MaxRating [get, set, inherited]

Maximum rating value.

Implements IRatingPredictor.

int MaxUserID [get, set, inherited]

Maximum user ID.

virtual float MinRating [get, set, inherited]

Minimum rating value.

Implements IRatingPredictor.

uint NumIter [get, set]

number of iterations over the dataset to perform

Implements IIterativeModel.

override IRatings Ratings [get, set, inherited]

The rating data.

Reimplemented from RatingPredictor.

float RegAlpha [get, set]

regularization for the user-wise alphas

float RegI [get, set]

regularization for the item bias

float RegItemBiasByTimeBin [get, set]

regularization for the bin-wise item bias

float RegU [get, set]

regularization for the user bias

float RegUserBiasByDay [get, set]

regularization for the day-wise user bias

float RegUserScaling [get, set]

regularization for the user scaling factor

float RegUserScalingByDay [get, set]

regularization for the day-wise user scaling factor

virtual ITimedRatings TimedRatings [get, set, inherited]

the rating data, including time information

Implements ITimeAwareRatingPredictor.

float UserBiasByDayLearnRate [get, set]

learn rate for the day-wise user bias

float UserBiasLearnRate [get, set]

learn rate for the user bias

float UserScalingByDayLearnRate [get, set]

learn rate for the day-wise user scaling factor

float UserScalingLearnRate [get, set]

learn rate for the user-wise scaling factor

The documentation for this class was generated from the following file:

TimeAwareBaseline.cs

Public Member Functions

Protected Member Functions

Protected Attributes

Properties

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation

Member Data Documentation

Property Documentation