Gaussian-Gamma collaborative filtering: A hierarchical Bayesian model for recommender systems
作者：
Highlights：
• The Gamma-Gaussian assumption on the ratings. It is a heavy tail distribution so the model is more robust.
• The Gamma-Gaussian assumption on the latent features. Hence we do not need to specify the regularization term manually.
• The Gibbs sampling of the parameters and the statistical explanation of the updating formulas.
摘要
•The Gamma-Gaussian assumption on the ratings. It is a heavy tail distribution so the model is more robust.•The Gamma-Gaussian assumption on the latent features. Hence we do not need to specify the regularization term manually.•The Gibbs sampling of the parameters and the statistical explanation of the updating formulas.
论文关键词：Gaussian-Gamma distribution,Recommender system,Hierarchical Bayesian model,Gibbs Sampling,Performance evaluation
论文评审过程：Received 28 September 2016, Revised 25 March 2017, Accepted 25 March 2017, Available online 27 April 2017, Version of Record 20 February 2019.
论文官网地址：https://doi.org/10.1016/j.jcss.2017.03.007
