Base class for IFactorizers, provides ID to index mapping

Factorizes the rating matrix using "Alternating-Least-Squares with Weighted-λ-Regularization" as described in "Large-scale Collaborative Filtering for the Netflix Prize" also supports the implicit feedback variant of this approach as described in "Collaborative Filtering for Implicit Feedback Datasets" available at http://research.yahoo.com/pub/2433

A factorization of the rating matrix

Provides a file-based persistent store.

A IPersistenceStrategy which does nothing.

Minimalistic implementation of Parallel SGD factorizer based on "Scalable Collaborative Filtering Approaches for Large Recommender Systems" and "Hogwild!: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent"

Matrix factorization with user and item biases for rating prediction, trained with plain vanilla SGD

SVD++, an enhancement of classical matrix factorization for rating prediction. Additionally to using ratings (how did people rate?) for learning, this model also takes into account who rated what. Yehuda Koren: Factorization Meets the Neighborhood: a Multifaceted Collaborative Filtering Model, KDD 2008. http://research.yahoo.com/files/kdd08koren.pdf

A IRecommender that uses matrix factorization (a projection of users and items onto a feature space)

Implementation must be able to create a factorization of a rating matrix

Provides storage for Factorizations