A new item-based deep network structure using a restricted Boltzmann machine for collaborative filtering

作     者：Yong-ping DU Chang-qing YAO Shu-hua HUO Jing-xuan LIU 

作者机构：Institute of Computer ScienceBeijing University of TechnologyBeijing 100124China Institute of Scientific and Technical information of ChinaBeijing 100038China 

出 版 物：《Frontiers of Information Technology & Electronic Engineering》 (信息与电子工程前沿（英文版）)

年 卷 期：2017年第18卷第5期

页      面：658-666页

核心收录：

学科分类：081203[工学-计算机应用技术] 08[工学] 0835[工学-软件工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Project supported by the National Science and Technology Suppor Plan(No.2013BAH21B02-01) the Beijing Natural Science Foundation(No.4153058) 

主　　题：Restricted Boltzmann machine Deep network structure Collaborative filtering Recommendation system 

摘      要：The collaborative filtering(CF) technique has been widely used recently in recommendation systems. It needs historical data to give predictions. However, the data sparsity problem still exists. We propose a new item-based restricted Boltzmann machine(RBM) approach for CF and use the deep multilayer RBM network structure, which alleviates the data sparsity problem and has excellent ability to extract features. Each item is treated as a single RBM, and different items share the same weights and biases. The parameters are learned layer by layer in the deep network. The batch gradient descent algorithm with minibatch is used to increase the convergence speed. The new feature vector discovered by the multilayer RBM network structure is very effective in predicting a rating and achieves a better result. Experimental results on the data set of MovieL ens show that the item-based multilayer RBM approach achieves the best performance, with a mean absolute error of 0.6424 and a root-mean-square error of 0.7843.