Entity representation for pairwise collaborative ranking using restricted Boltzmann machine
作者：
Highlights：
• RBMRank is a novel model-based framework for pairwise collaborative ranking
• RBMRank uses RBM to individually learn representations of users and preferences
• RBMRank suggests a link prediction model in signed networks for opinion prediction.
• RBMRank can simply update entities representation without model reconstruction
• RBMRank improves NDCG10 up to 10% compared to other collaborative ranking methods.
摘要
•RBMRank is a novel model-based framework for pairwise collaborative ranking•RBMRank uses RBM to individually learn representations of users and preferences•RBMRank suggests a link prediction model in signed networks for opinion prediction.•RBMRank can simply update entities representation without model reconstruction•RBMRank improves NDCG10 up to 10% compared to other collaborative ranking methods.
论文关键词：Deep learning,Restricted Boltzmann machine,Collaborative filtering,Pairwise preferences
论文评审过程：Received 24 January 2018, Revised 15 August 2018, Accepted 6 September 2018, Available online 7 September 2018, Version of Record 18 September 2018.
论文官网地址：https://doi.org/10.1016/j.eswa.2018.09.013
