C# (CSharp) MyMediaLite.RatingPrediction Namespace

Classes

Name Description
BiPolarSlopeOne Bi-polar frequency-weighted Slope-One rating prediction
BiasedMatrixFactorization Matrix factorization with explicit user and item bias, learning is performed by stochastic gradient descent
EntityAverage Abstract class that uses an average (by entity) rating value for predictions
FactorWiseMatrixFactorization Matrix factorization with factor-wise learning
GlobalAverage Uses the average rating value over all ratings for prediction
ItemAttributeKNN Attribute-aware weighted item-based kNN recommender
ItemAverage Uses the average rating value of an item for prediction
ItemKNN Weighted item-based kNN
ItemKNNCosine Weighted item-based kNN with cosine similarity
ItemKNNPearson Weighted item-based kNN with pearson correlation
KNN Base class for rating predictors that use some kind of kNN
LogisticRegressionMatrixFactorization Recommender that performs a matrix factorization optimized for regularized logit loss
MatrixFactorization Simple matrix factorization class, learning is performed by stochastic gradient descent (SGD)
NewKNN kNN-based rating predictors (not working yet)
Prediction Class that contains static methods for rating prediction
RatingPredictor Abstract class for rating predictors that keep the rating data in memory for training (and possibly prediction)
SimpleMatrixFactorization Simple matrix factorization class
SlopeOne Frequency-weighted Slope-One rating prediction
SocialMF Social-network-aware matrix factorization
UserAttributeKNN Weighted kNN recommender based on user attributes
UserAverage Uses the average rating value of a user for predictions
UserItemBaseline Baseline method for rating prediction
UserKNN Weighted user-based kNN
UserKNNCosine Weighted user-based kNN with cosine similarity
UserKNNPearson Weighted user-based kNN with Pearson correlation