TY - JOUR
T1 - An inherently nonnegative latent factor model for high-dimensional and sparse matrices from industrial applications
AU - Luo, Xin
AU - Zhou, Mengchu
AU - Li, Shuai
AU - Shang, Mingsheng
N1 - Funding Information:
Manuscript received July 30, 2017; revised September 22, 2017; accepted October 8, 2017. Date of publication October 27, 2017; date of current version May 2, 2018. This work was supported in part by the National Key Research and Development Program of China under Grant 2017YFC0804002, in part by the International Joint Project funded jointly by the Royal Society of the UK and the National Natural Science Foundation of China under Grant 61611130209, in part by the National Natural Science Foundation of China under Grant 61772493, Grant 91646114, and Grant 51609229, in part by FDCT (Fundo para o Desenvolvimento das Ciencias e da Tecnologia) under Grant 119/2014/A3, in part by the Pioneer Hundred Talents Program of Chinese Academy of Sciences, and in part by the Young Scientist Foundation of Chongqing under Grant cstc2014kjrc-qnrc40005. Paper no. TII-17-1697. (Xin Luo and MengChu Zhou contributed equally to this work. Corresponding authors: Xin Luo and Shuai Li.) X. Luo and M. S. Shang are with Chongqing Key Laboratory of Big Data and Intelligent Computing, Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing 400714, China (e-mail: luoxin21@cigit.ac.cn; msshang@cigit.ac.cn).
Publisher Copyright:
© 2005-2012 IEEE.
PY - 2018/5
Y1 - 2018/5
N2 - High-dimensional and sparse (HiDS) matrices are commonly encountered in many big-data-related and industrial applications like recommender systems. When acquiring useful patterns from them, nonnegative matrix factorization (NMF) models have proven to be highly effective owing to their fine representativeness of the nonnegative data. However, current NMF techniques suffer from: 1) inefficiency in addressing HiDS matrices; and 2) constraints in their training schemes. To address these issues, this paper proposes to extract nonnegative latent factors (NLFs) from HiDS matrices via a novel inherently NLF (INLF) model. It bridges the output factors and decision variables via a single-element-dependent mapping function, thereby making the parameter training unconstrained and compatible with general training schemes on the premise of maintaining the nonnegativity constraints. Experimental results on six HiDS matrices arising from industrial applications indicate that INLF is able to acquire NLFs from them more efficiently than any existing method does.
AB - High-dimensional and sparse (HiDS) matrices are commonly encountered in many big-data-related and industrial applications like recommender systems. When acquiring useful patterns from them, nonnegative matrix factorization (NMF) models have proven to be highly effective owing to their fine representativeness of the nonnegative data. However, current NMF techniques suffer from: 1) inefficiency in addressing HiDS matrices; and 2) constraints in their training schemes. To address these issues, this paper proposes to extract nonnegative latent factors (NLFs) from HiDS matrices via a novel inherently NLF (INLF) model. It bridges the output factors and decision variables via a single-element-dependent mapping function, thereby making the parameter training unconstrained and compatible with general training schemes on the premise of maintaining the nonnegativity constraints. Experimental results on six HiDS matrices arising from industrial applications indicate that INLF is able to acquire NLFs from them more efficiently than any existing method does.
KW - Big data
KW - high-dimensional and sparse matrix
KW - learning algorithms
KW - missing-data estimation
KW - nonnegative latent factor analysis
KW - optimization methods recommender system
UR - http://www.scopus.com/inward/record.url?scp=85032731287&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85032731287&partnerID=8YFLogxK
U2 - 10.1109/TII.2017.2766528
DO - 10.1109/TII.2017.2766528
M3 - Article
AN - SCOPUS:85032731287
SN - 1551-3203
VL - 14
SP - 2011
EP - 2022
JO - IEEE Transactions on Industrial Informatics
JF - IEEE Transactions on Industrial Informatics
IS - 5
ER -