TY - JOUR
T1 - Adaptive Weighted Sparse Principal Component Analysis for Robust Unsupervised Feature Selection
AU - Yi, Shuangyan
AU - He, Zhenyu
AU - Jing, Xiao Yuan
AU - Li, Yi
AU - Cheung, Yiu Ming
AU - Nie, Feiping
N1 - Funding Information:
Manuscript received February 4, 2018; revised June 12, 2018, September 29, 2018, January 26, 2019, and June 12, 2019; accepted July 5, 2019. Date of publication August 28, 2019; date of current version June 2, 2020. This work was supported in part by the National Natural Science Foundation of China under Grant 61672183 and Grant 61772141, in part by the Shenzhen Research Council under Grant JCYJ20170413104556946, Grant JCYJ20160406161948211, Grant JCYJ20160226201453085, and Grant KJYY20170724152625446, in part by the Natural Science Foundation of Guangdong Province under Grant 2015A030313544, in part by the Guangdong Provincial Natural Science Foundation under Grant 17ZK0422, and in part by the Guangzhou Science and Technology Planning Project under Grant201804010347. (Corresponding authors: Zhenyu He; Xiao-Yuan Jing.) S. Yi is with the Institute of Information Technology, Shenzhen Institute of Information Technology, Shenzhen 518172, China, and also with the School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen 518055, China (e-mail: [email protected]).
PY - 2020/6
Y1 - 2020/6
N2 - Current unsupervised feature selection methods cannot well select the effective features from the corrupted data. To this end, we propose a robust unsupervised feature selection method under the robust principal component analysis (PCA) reconstruction criterion, which is named the adaptive weighted sparse PCA (AW-SPCA). In the proposed method, both the regularization term and the reconstruction error term are constrained by the $\ell _{2,1}$ -norm: the $\ell _{2,1}$ -norm regularization term plays a role in the feature selection, while the $\ell _{2,1}$ -norm reconstruction error term plays a role in the robust reconstruction. The proposed method is in a convex formulation, and the selected features by it can be used for robust reconstruction and clustering. Experimental results demonstrate that the proposed method can obtain better reconstruction and clustering performance, especially for the corrupted data.
AB - Current unsupervised feature selection methods cannot well select the effective features from the corrupted data. To this end, we propose a robust unsupervised feature selection method under the robust principal component analysis (PCA) reconstruction criterion, which is named the adaptive weighted sparse PCA (AW-SPCA). In the proposed method, both the regularization term and the reconstruction error term are constrained by the $\ell _{2,1}$ -norm: the $\ell _{2,1}$ -norm regularization term plays a role in the feature selection, while the $\ell _{2,1}$ -norm reconstruction error term plays a role in the robust reconstruction. The proposed method is in a convex formulation, and the selected features by it can be used for robust reconstruction and clustering. Experimental results demonstrate that the proposed method can obtain better reconstruction and clustering performance, especially for the corrupted data.
KW - clustering
KW - feature selection
KW - l2,1-norm
KW - reconstruction
UR - http://www.scopus.com/inward/record.url?scp=85085903991&partnerID=8YFLogxK
U2 - 10.1109/TNNLS.2019.2928755
DO - 10.1109/TNNLS.2019.2928755
M3 - Journal article
C2 - 31478875
AN - SCOPUS:85085903991
SN - 2162-237X
VL - 31
SP - 2153
EP - 2163
JO - IEEE Transactions on Neural Networks and Learning Systems
JF - IEEE Transactions on Neural Networks and Learning Systems
IS - 6
M1 - 8818654
ER -