CF4J 2.0: Adapting Collaborative Filtering for Java to new challenges of collaborative filtering based recommender systems

作者:

Highlights:

摘要

CF4J 2.0 is a framework for conducting research experiments based on collaborative filtering. This framework has been designed keeping the scientific community in mind. It includes major features such as a high number of implemented algorithms from the state-of-the-art, several quality measures and parallel execution of the techniques, as well as abstract classes and interfaces to allow developers to extend and customize the library. Furthermore, this new version of the library focuses on the following key features: simple deployment of collaborative filtering experiments, reproducible science, hyper-parameter optimization, data analysis, and openness to the community as an open-source project.

论文关键词:Collaborative filtering,Recommender system,Matrix Factorization

论文评审过程:Received 30 April 2020, Revised 23 November 2020, Accepted 24 November 2020, Available online 25 November 2020, Version of Record 19 January 2021.

论文官网地址:https://doi.org/10.1016/j.knosys.2020.106629