MyMediaLite  3.11
Classes
Package MyMediaLite.ItemRecommendation

This namespace contains item recommenders and some helper classes for item recommendation. More...

Classes

class  BPRMF
 Matrix factorization model for item prediction (ranking) optimized for BPR More...
 
class  BPRSLIM
 Sparse Linear Methods (SLIM) for item prediction (ranking) optimized for BPR-Opt optimization criterion More...
 
class  Extensions
 Class that contains static methods for item prediction More...
 
class  ExternalItemRecommender
 Uses externally computed predictions More...
 
interface  IFoldInItemRecommender
 Item recommender that allows folding in new users More...
 
interface  IIncrementalItemRecommender
 Interface for item recommenders More...
 
class  IncrementalItemRecommender
 Base class for item recommenders that support incremental updates More...
 
class  ItemAttributeKNN
 k-nearest neighbor (kNN) item-based collaborative filtering using the correlation of the item attibutes More...
 
class  ItemKNN
 k-nearest neighbor (kNN) item-based collaborative filtering More...
 
class  ItemRecommender
 Abstract item recommender class that loads the (positive-only implicit feedback) training data into memory and provides flexible access to it. More...
 
interface  ITransductiveItemRecommender
 Interface for item recommenders that take into account some test data for training More...
 
class  KNN
 Base class for item recommenders that use some kind of k-nearest neighbors (kNN) model More...
 
class  LeastSquareSLIM
 Sparse Linear Methods (SLIM) for item prediction (ranking) optimized for the elastic net loss More...
 
class  MF
 Abstract class for matrix factorization based item predictors More...
 
class  MostPopular
 Most-popular item recommender More...
 
class  MostPopularByAttributes
 Recommend most popular items by attribute More...
 
class  MultiCoreBPRMF
 Matrix factorization for BPR on multiple cores More...
 
class  Random
 Random item recommender for use as experimental baseline More...
 
class  SLIM
 Abstract class for SLIM based item predictors proposed by Ning and Karypis More...
 
class  SoftMarginRankingMF
 Matrix factorization model for item prediction optimized for a soft margin (hinge) ranking loss, using stochastic gradient descent (as in BPR-MF). More...
 
class  UserAttributeKNN
 k-nearest neighbor (kNN) user-based collaborative filtering using the correlation of the user attibutes More...
 
class  UserKNN
 k-nearest neighbor user-based collaborative filtering More...
 
class  WeightedBPRMF
 Weigthed BPR-MF with frequency-adjusted sampling More...
 
class  WRMF
 Weighted matrix factorization method proposed by Hu et al. and Pan et al. More...
 
class  Zero
 Constant item recommender for use as experimental baseline. Always predicts a score of zero More...
 

Detailed Description

This namespace contains item recommenders and some helper classes for item recommendation.