Dynamic graph learning for spectral feature selection

被引:79
作者
Zheng, Wei [1 ]
Zhu, Xiaofeng [1 ]
Zhu, Yonghua [2 ]
Hu, Rongyao [1 ]
Lei, Cong [1 ]
机构
[1] Guangxi Normal Univ, Guangxi Key Lab Multisource Informat Min & Secur, Guilin 541004, Guangxi, Peoples R China
[2] Guangxi Univ, Sch Comp Elect & Informat, Nanning 530004, Guangxi, Peoples R China
关键词
Graph learning; Optimization; Spectral feature selection; SUPPORT VECTOR MACHINES; ASSOCIATION RULES; CLASSIFICATION; REGRESSION; ALGORITHM;
D O I
10.1007/s11042-017-5272-y
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Previous spectral feature selection methods generate the similarity graph via ignoring the negative effect of noise and redundancy of the original feature space, and ignoring the association between graph matrix learning and feature selection, so that easily producing suboptimal results. To address these issues, this paper joints graph learning and feature selection in a framework to obtain optimal selected performance. More specifically, we use the least square loss function and an l(2,1)-norm regularization to remove the effect of noisy and redundancy features, and use the resulting local correlations among the features to dynamically learn a graph matrix from a low-dimensional space of original data. Experimental results on real data sets show that our method outperforms the state-of-the-art feature selection methods for classification tasks.
引用
收藏
页码:29739 / 29755
页数:17
相关论文
共 55 条
[1]  
[Anonymous], IEEE INT C DAT MIN
[2]  
[Anonymous], UNSUPERVISED SPECT C
[3]  
[Anonymous], DATA PREPARATION DAT
[4]  
[Anonymous], 2003, NIPS
[5]  
[Anonymous], WORKSH SPEECH NAT LA
[6]  
Boyd S., 2006, IEEE Trans. Autom. Control, V51, P1859, DOI DOI 10.1109/TAC.2006.884922
[7]  
Cai D, 2010, P 16 ACM SIGKDD INT, P333, DOI DOI 10.1145/1835804.1835848
[8]   Iteratively Reweighted Least Squares Minimization for Sparse Recovery [J].
Daubechies, Ingrid ;
Devore, Ronald ;
Fornasier, Massimo ;
Guentuerk, C. Sinan .
COMMUNICATIONS ON PURE AND APPLIED MATHEMATICS, 2010, 63 (01) :1-38
[9]  
De Wang, 2014, Machine Learning and Knowledge Discovery in Databases. European Conference, ECML PKDD 2014. Proceedings: LNCS 8726, P306, DOI 10.1007/978-3-662-44845-8_20
[10]   A new approximate maximal margin classification algorithm [J].
Gentile, C .
JOURNAL OF MACHINE LEARNING RESEARCH, 2002, 2 (02) :213-242