An enhanced matrix completion method based on non-negative latent factors for recommendation system
作者：
Highlights：
• A new non-negative latent factor model is proposed.
• Various non-negative latent factors are taken into consideration.
• We adopt a momentum additive gradient descent method to accelerate the learning.
• A truncating strategy is used to guarantee the non-negativity of latent factors.
• Empirical studies on public datasets demonstrate the effectiveness of our models.
摘要
•A new non-negative latent factor model is proposed.•Various non-negative latent factors are taken into consideration.•We adopt a momentum additive gradient descent method to accelerate the learning.•A truncating strategy is used to guarantee the non-negativity of latent factors.•Empirical studies on public datasets demonstrate the effectiveness of our models.
论文关键词：Model-based collaborative filtering,Matrix factorization,Non-negative latent factor,Additive gradient descent algorithm,Generalized momentum method
论文评审过程：Received 17 December 2020, Revised 18 October 2021, Accepted 23 March 2022, Available online 2 April 2022, Version of Record 12 April 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.116985
