In memory data storage
Information required by Collaborative Filtering (i.e. users, items and ratings) is stored in memory using efficient data structures.
Extensible pipeline
Collaborative Filtering execution pipeline can be easily extended using the abstract classes and methods provided.
Benchmark datasets
Most popular datasets used in Collaborative Filtering research has been preloaded.
Hyperparameter optimization
Models' hyperparameter can be quickly tuned using the included grid search utility.
Data and results analysis
Experimental results can be efficiently analyzed using the amazing plots.
State of the Art baselines
Some of the most referenced recommendations models has been included.
Compiles and runs with Java 8
Fully support of Java 8.
Open to the community
As an open source project, the repository in
GitHub is fully available for other developers to extend and improve the library.