MyMediaLite  3.03
Classes | Enumerations
Package MyMediaLite.RatingPrediction

This namespace contains rating predictors and some helper classes for rating prediction. More...

Classes

class  BiasedMatrixFactorization
 Matrix factorization with explicit user and item bias, learning is performed by stochastic gradient descent. More...
class  BiPolarSlopeOne
 Bi-polar frequency-weighted Slope-One rating prediction. More...
class  CoClustering
 Co-clustering for rating prediction. More...
class  Constant
 Uses a constant rating value for prediction. More...
class  EntityAverage
 Abstract class that uses an average (by entity) rating value for predictions. More...
class  Extensions
 Class that contains static methods for rating prediction. More...
class  FactorWiseMatrixFactorization
 Matrix factorization with factor-wise learning. More...
class  FoldInRatingPredictorExtensions
 Extension methods for IFoldInRatingPredictor. More...
class  GlobalAverage
 Uses the average rating value over all ratings for prediction. More...
interface  IFoldInRatingPredictor
 Rating predictor that allows folding in new users. More...
interface  IIncrementalRatingPredictor
 Interface for rating predictors which support incremental training. More...
class  IncrementalRatingPredictor
 Base class for rating predictors that support incremental training. More...
interface  IRatingPredictor
 Interface for rating predictors. More...
class  ItemAttributeKNN
 Attribute-aware weighted item-based kNN recommender. More...
class  ItemAverage
 Uses the average rating value of an item for prediction. More...
class  ItemKNN
 Weighted item-based kNN. More...
interface  ITimeAwareRatingPredictor
 Interface for time-aware rating predictors. More...
interface  ITransductiveRatingPredictor
 Rating predictor that knows beforehand what it will have to rate. More...
class  KNN
 Base class for rating predictors that use some kind of kNN. More...
class  LatentFeatureLogLinearModel
 Latent-feature log linear model. More...
class  MatrixFactorization
 Simple matrix factorization class, learning is performed by stochastic gradient descent. More...
class  NaiveBayes
 Attribute-aware rating predictor using Naive Bayes. More...
class  Random
 Uses a random rating value for prediction. More...
class  RatingPredictor
 Abstract class for rating predictors that keep the rating data in memory for training (and possibly prediction) More...
class  SigmoidCombinedAsymmetricFactorModel
 Asymmetric factor model which represents items in terms of the users that rated them, and users in terms of the items they rated. More...
class  SigmoidItemAsymmetricFactorModel
 Asymmetric factor model. More...
class  SigmoidSVDPlusPlus
 SVD++: Matrix factorization that also takes into account _what_ users have rated; variant that uses a sigmoid function. More...
class  SigmoidUserAsymmetricFactorModel
 Asymmetric factor model which represents items in terms of the users that rated them. More...
class  SlopeOne
 Frequency-weighted Slope-One rating prediction. More...
class  SocialMF
 Social-network-aware matrix factorization. More...
class  SVDPlusPlus
 SVD++: Matrix factorization that also takes into account _what_ users have rated. More...
class  TimeAwareBaseline
 Time-aware bias model. More...
class  TimeAwareBaselineWithFrequencies
 Time-aware bias model with frequencies. More...
class  TimeAwareRatingPredictor
 Abstract class for time-aware rating predictors. More...
class  TransductiveRatingPredictorExtensions
 Helper methods for ITransductiveRatingPredictor. More...
class  UserAttributeKNN
 Weighted kNN recommender based on user attributes. More...
class  UserAverage
 Uses the average rating value of a user for predictions. More...
class  UserItemBaseline
 baseline method for rating prediction More...
class  UserKNN
 Weighted user-based kNN. More...

Enumerations

enum  OptimizationTarget { RMSE, MAE, LogisticLoss }
 Enumeration to represent different optimization targets for rating prediction. More...

Detailed Description

This namespace contains rating predictors and some helper classes for rating prediction.


Enumeration Type Documentation

Enumeration to represent different optimization targets for rating prediction.

Enumerator:
RMSE 

(root) mean square error

MAE 

mean absolute error

LogisticLoss 

log likelihood of the data (as in logistic regression)