TY - JOUR
T1 - A deep latent factor model for high-dimensional and sparse matrices in recommender systems
AU - Wu, Di
AU - Luo, Xin
AU - Shang, Mingsheng
AU - He, Yi
AU - Wang, Guoyin
AU - Zhou, Mengchu
N1 - Funding Information:
Manuscript received May 31, 2019; accepted July 19, 2019. Date of publication August 15, 2019; date of current version June 16, 2021. This work was supported in part by the National Key Research and Development Program of China under Grant 2017YFC0804002, in part by the National Natural Science Foundation of China under Grant 61702475, Grant 61772493, and Grant 91646114, in part by the Chongqing Basic Research and Frontier Exploration under Grant cstc2019jcyj-msxm1750, in part by the Chongqing Overseas Scholars Innovation Program under Grant cx2017012 and Grant cx2018011, in part by the Chongqing Research Program of Key Standard Technologies Innovation of Key Industries under Grant cstc2017zdcy-zdyfX0076 and Grant cstc2018jszx-cyztzxX0025, in part by the Chongqing Research Program of Technology Innovation and Application under Grant cstc2017rgzn-zdyfX0020, Grant cstc2017zdcy-zdyf0554, and Grant cstc2017rgzn-zdyf0118, and in part by the Pioneer Hundred Talents Program of Chinese Academy of Sciences. This paper was recommended by Associate Editor X. Wang. (Di Wu and Mingsheng Shang are co-first authors of this paper.) (Corresponding author: Xin Luo.) D. Wu is with the Chongqing Engineering Research Center of Big Data Application for Smart Cities, Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing 400714, China, also with the Chongqing Key Laboratory of Big Data and Intelligent Computing, Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing 400714, China, and also with the School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing 100049, China (e-mail: wudi@cigit.ac.cn).
Publisher Copyright:
© 2013 IEEE.
PY - 2021/7
Y1 - 2021/7
N2 - Recommender systems (RSs) commonly adopt a user-item rating matrix to describe users' preferences on items. With users and items exploding, such a matrix is usually high-dimensional and sparse (HiDS). Recently, the idea of deep learning has been applied to RSs. However, current deep-structured RSs suffer from high computational complexity. Enlightened by the idea of deep forest, this paper proposes a deep latent factor model (DLFM) for building a deep-structured RS on an HiDS matrix efficiently. Its main idea is to construct a deep-structured model by sequentially connecting multiple latent factor (LF) models instead of multilayered neural networks through a nonlinear activation function. Thus, the computational complexity grows linearly with its layer count, which is easy to resolve in practice. The experimental results on four HiDS matrices from industrial RSs demonstrate that when compared with state-of-the-art LF models and deep-structured RSs, DLFM can well balance the prediction accuracy and computational efficiency, which well fits the desire of industrial RSs for fast and right recommendations.
AB - Recommender systems (RSs) commonly adopt a user-item rating matrix to describe users' preferences on items. With users and items exploding, such a matrix is usually high-dimensional and sparse (HiDS). Recently, the idea of deep learning has been applied to RSs. However, current deep-structured RSs suffer from high computational complexity. Enlightened by the idea of deep forest, this paper proposes a deep latent factor model (DLFM) for building a deep-structured RS on an HiDS matrix efficiently. Its main idea is to construct a deep-structured model by sequentially connecting multiple latent factor (LF) models instead of multilayered neural networks through a nonlinear activation function. Thus, the computational complexity grows linearly with its layer count, which is easy to resolve in practice. The experimental results on four HiDS matrices from industrial RSs demonstrate that when compared with state-of-the-art LF models and deep-structured RSs, DLFM can well balance the prediction accuracy and computational efficiency, which well fits the desire of industrial RSs for fast and right recommendations.
KW - Big data
KW - deep model
KW - high-dimensional and sparse (HiDS) matrix
KW - latent factor (LF) analysis
KW - recommender system (RS)
UR - http://www.scopus.com/inward/record.url?scp=85112187060&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85112187060&partnerID=8YFLogxK
U2 - 10.1109/TSMC.2019.2931393
DO - 10.1109/TSMC.2019.2931393
M3 - Article
AN - SCOPUS:85112187060
SN - 2168-2216
VL - 51
SP - 4285
EP - 4296
JO - IEEE Transactions on Systems, Man, and Cybernetics: Systems
JF - IEEE Transactions on Systems, Man, and Cybernetics: Systems
IS - 7
M1 - 8802269
ER -