TY - GEN
T1 - Efficient extraction of non-negative latent factors from high-dimensional and sparse matrices in industrial applications
AU - Luo, Xin
AU - Shang, Mingsheng
AU - Li, Shuai
PY - 2017/1/31
Y1 - 2017/1/31
N2 - High-dimensional and sparse (HiDS) matrices are commonly encountered in many big data-related industrial applications like recommender systems. When acquiring useful patterns from them, non-negative matrix factorization (NMF) models have proven to be highly effective because of their fine representativeness of non-negative data. However, current NMF techniques suffer from a) inefficiency in addressing HiDS matrices, and b) constrained training schemes lack of flexibility, extensibility and adaptability. To address these issues, this work proposes to factorize industrial-size sparse matrices via a novel Inherently Non-negative Latent Factor (INLF) model. It connects the output factors and decision variables via a single-element-dependent sigmoid function, thereby innovatively removing the non-negativity constraints from its training process without impacting the solution accuracy. Hence, its training process is unconstrained, highly flexible and compatible with general learning schemes. Experimental results on five HiDS matrices generated by industrial applications indicate that INLF is able to acquire non-negative latent factors from them in a more efficient manner than any existing method does.
AB - High-dimensional and sparse (HiDS) matrices are commonly encountered in many big data-related industrial applications like recommender systems. When acquiring useful patterns from them, non-negative matrix factorization (NMF) models have proven to be highly effective because of their fine representativeness of non-negative data. However, current NMF techniques suffer from a) inefficiency in addressing HiDS matrices, and b) constrained training schemes lack of flexibility, extensibility and adaptability. To address these issues, this work proposes to factorize industrial-size sparse matrices via a novel Inherently Non-negative Latent Factor (INLF) model. It connects the output factors and decision variables via a single-element-dependent sigmoid function, thereby innovatively removing the non-negativity constraints from its training process without impacting the solution accuracy. Hence, its training process is unconstrained, highly flexible and compatible with general learning schemes. Experimental results on five HiDS matrices generated by industrial applications indicate that INLF is able to acquire non-negative latent factors from them in a more efficient manner than any existing method does.
KW - Big data
KW - High-dimensional and sparse matrices
KW - Inherently non-negative
KW - Latent factor
KW - Missing-data estimation
KW - Non-negative factorization
KW - Unconstrained
UR - http://www.scopus.com/inward/record.url?scp=85014567655&partnerID=8YFLogxK
U2 - 10.1109/ICDM.2016.58
DO - 10.1109/ICDM.2016.58
M3 - Conference article published in proceeding or book
AN - SCOPUS:85014567655
T3 - Proceedings - IEEE International Conference on Data Mining, ICDM
SP - 311
EP - 319
BT - Proceedings - 16th IEEE International Conference on Data Mining, ICDM 2016
A2 - Bonchi, Francesco
A2 - Wu, Xindong
A2 - Baeza-Yates, Ricardo
A2 - Domingo-Ferrer, Josep
A2 - Zhou, Zhi-Hua
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 16th IEEE International Conference on Data Mining, ICDM 2016
Y2 - 12 December 2016 through 15 December 2016
ER -