Abstract
since amount of unlabeled and high-dimensional datasets need to be preprocessed, unsupervised learning plays a more and more important role in machine learning field. This paper proposed a novel unsupervised feature selection algorithm that can select informative features from dataset without label, by mixing two sparse representation and self-representation loss function into a unified framework. That is, we use self-representation loss function to represent every feature with remainder features and achieve minimum reconstruction mirror, and then utilize l 2 , 1-norm regularization term and l 1-norm regularization term simultaneously to enforce coefficient matrix to be sparse, such that filter redundant and irrelative features in order to conduct feature selection, where l 2 , 1-norm regularization can enforce group sparsity while l 1-norm regularization enforce element sparsity. By this way that utilize both of sparse representation terms, we can choose representative features more accurately. At final, we feed reduced data into support vector machine (SVM) to conduct classification accuracy, which is main assessment criteria to validate performance of algorithm. Extensive experiments on synthetic datasets and real-world datasets have exhibited that our proposed method outperform most of common-used methods, such as PCA, LPP and so on.
Similar content being viewed by others
References
Cheng B, Zhang D, Chen S, Kaufer D, Shen D (2013) Semi-supervised multimodal relevance vector regression improves cognitive performance estimation from imaging and biological biomarkers. Neuroinformatics 11(3):339–353
Cheng D, Zhang S, Liu X et al. (2015) Feature selection by combining subspace learning with sparse representation[J], Multimedia Systems., 1–7
Feng Y, Xiao J, Zhuang Y et al. (2012) Adaptive Unsupervised Multi-view Feature Selection for Visual Concept Recognition [J], Compu Vis – ECCV 2012, 7724:343–357
Hai TN, Franke K, Petrovic S (2011) On General Definition of L1-norm Support Vector Machines for Feature Selection [J]. 1(2):279–283
He R, Tan T, Wang L, et al. (2012) l2, 1 Regularized correntropy for robust feature selection [C]. IEEE Conference on Computer Vision & Pattern Recognition. 2504–2511
Jolliffe IT (1986) Principal Component. Springer Series in Statistics Analysis. Springer, New York. doi:10.1007/b98835
Kan M, Shan S, Zhang H (2012) Multi-view discriminant analysis [J]. Comput Vis ECCV 2012(7525):808–821
Lai H, Pan Y, Liu C et al (2013) Sparse learning-to-rank via an efficient primal-dual algorithm [J]. IEEE Trans Comput 62(6):1221–1233
Laporte L, Flamary R, Canu S et al (2015) Non-convex Regularizations for Feature Selection in Ranking With Sparse SVM [J]. IEEE Trans Neural Netw Learn Syst 25(6):1118–1130
Leordeanu M, Hebert M (2009) Unsupervised learning for graph matching [J]. Int J Comput Vis 96(1):28–45
Lu H, Plataniotis KN, Venetsanopoulos AN (2011) A survey of multilinear subspace learning for tensor data [J]. Pattern Recogn 44(7):1540–1551
Nie F, Huang H, Cai X, Ding CHQ (2010) Efficient and robust feature selection via joint l_2,1-norms minimization [J], in: NIPS, 1813–1821
Peng H, Long F, Ding C (2005) Feature selection based on mutual information: criteria of max-dependency, max-relevance, and Min-redundancy [J]. IEEE Trans Pattern Anal Mach Intell 27(8):1226–1238
Qian M, Zhai C (2013) Robust unsupervised feature selection [C], IJCAI ‘13 proceedings of the twenty-third international joint conference on artificial intelligence, 1621–1627
Qin Y, Zhang S, Zhu X et al (2007) Semi-parametric optimization for missing data imputation [J]. Appl Intell 27(1):79–88
Ryali S, Menon V (2009) Feature Selection and Classification of fMRI data using Logistic Regression with L1 norm regularization [J]. Neuroimage 47(47):S57
Shakhnarovich G, Moghaddam B (2004) Face recognition in subspaces, In: S.Z. Li, A.K. Jain (Eds.), Handbook of Face Recognition, Springer-Verlag, 141–168
Shang R, Zhang Z, Jiao L et al (2016) Self-representation based dual-graph regularized feature selection clustering [J]. Neurocomputing 171(C):1242–1253
Stonnington CM, Chu C et al (2010) Predicting clinical scores from magnetic resonance scans in Alzheimer’s disease. NeuroImage 51(4):1405–1413
Tomar D, Agarwal S (2014) Feature selection based Least Square twin support vector machine for diagnosis of heart disease [J]. Int J Biol Sci Bio/Technology 6:69–82
Wang JY, Bensmail H, Gao X (2014) Feature selection and multi-kernel learning for sparse representation on a manifold [J]. Neural Netw 51(3):9–16
Wang S, Lu J, Gu X, Du H et al (2016) Semi-supervised linear discriminant analysis for dimension reduction and classification [J]. Pattern Recogn 57:179–189
Weston J, Mukherjee S, Chapelle O et al (2000) Feature selection for SVMs [J]. Adv Neural Inf Proces Syst 13:668–674
Xiao R, Zhao Q (2011) David Zhang et al. facial expression recognition on multiple manifolds [J]. Pattern Recogn 44(1):107–116
Xu Y, Wang C, Lai J (2016) Weighted multi-view clustering with feature selection [J]. Pattern Recogn 53:25–35
Yang Y, Shen HT, Ma Z, et al. (2011) l 2,1-norm regularized discriminative feature selection for unsupervised learning [C]. International Joint Conference on Artificial Intelligence. 1589–1594
Yuan GX, Ho CH, Lin CJ (2011) An improved GLMNET for L1-regularized logistic regression [J]. J Mach Learn Res 13(1):33–41
Zhang C, Qin Y, Zhu X et al. (2006) Clustering-based Missing Value Imputation for Data Preprocessing [J]. pp. 1081–1086
Zhu Y, Lucey S (2013) Convolutional sparse coding for trajectory reconstruction [J]. IEEE Trans Pattern Anal Mach Intell 37(3):529–540
Zhu X, Zhang S, Zhang J, et al. (2007) Cost-Sensitive Imputing Missing Values with Ordering [C]. AAAI Conference on Artificial Intelligence, July 22–26, 2007, Vancouver, British Columbia, pp. 1922-1923
Zhu X, Zhang S, Jin Z, Zhang Z, Xu Z (2011a) Missing value estimation for mixed-attribute data sets. IEEE Trans Knowl Data Eng 23(1):110–121
Zhu Y, Cox M, Lucey S (2011b) 3D motion reconstruction for real-world camera motion [C], Cvpr, IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 1–8
Zhu X, Huang Z, Cui J, Shen H (2013a) Video-to-shot tag propagation by graph sparse group lasso. IEEE Trans Multimedia 15(3):633–646
Zhu X, Huang Z, Cheng H, Cui J, Shen H (2013b) Sparse hashing for fast multimedia search [J]. ACM Trans Inf Syst 31(2):9
Zhu X, Huang Z, Yang Y, Shen HT, Xu C, Luo J (2013c) Self-taught dimensionality reduction on the high-dimensional small-sized data [J]. Pattern Recogn 46(1):215–229
Zhu X, Suk H, Shen D (2014a) A novel matrix-similarity based loss function for joint regression and classification in AD diagnosis [J]. NeuroImage 100:91–105
Zhu X, Zhang L, Huang Z (2014b) A sparse embedding and least variance encoding approach to hashing. IEEE Trans Image Process 23(9):3737–3750
Zhu P, Zuo W, Zhang L et al (2015a) Unsupervised feature selection by regularized self-representation [J]. Pattern Recogn 48(2):438–446
Zhu X, Suk H, Shen D (2015b) Low-rank dimensionality reduction for multi-modality AD classification [J]. Neuroimage 100:91–105
Zhu Y, Huang D, De La Torre F, Lucey S (2015c) Complex non-rigid motion 3d reconstruction by union of subspaces [C]. 2014 I.E. Conf Comput Vis Pattern Recognit 37(3):1542–1549
Zhu X, Li X, Zhang S, et al. (2016a) Robust Joint Graph Sparse Coding for Unsupervised Spectral Feature Selection. [J]. IEEE Transactions on Neural Networks & Learning Systems, pp. 1–13
Zhu X, Li X, Zhang S (2016b) Block-Row Sparse Multiview Multilabel Learning for Image Classification [J]. IEEE Trans Cybern 46(2):1
Acknowledgments
This work was supported in part by the China “1000-Plan” National Distinguished Professorship; the Nation Natural Science Foundation of China (Grants No: 61263035, 61573270 and 61672177), the China 973 Program (Grant No: 2013CB329404); the China Key Research Program (Grant No: 2016YFB1000905); the Guangxi Natural Science Foundation (Grant No: 2015GXNSFCB139011); the Innovation Project of Guangxi Graduate Education under grant YCSZ2016046; the Guangxi High Institutions’ Program of Introducing 100 High-Level Overseas Talents; and the Guangxi Collaborative Innovation Center of Multi-Source Information Integration and Intelligent Processing; and the Guangxi “Bagui” Teams for Innovation and Research.
Author information
Authors and Affiliations
Corresponding author
Additional information
Rights and permissions
About this article
Cite this article
Zhu, Y., Zhang, X., Wen, G. et al. Double sparse-representation feature selection algorithm for classification. Multimed Tools Appl 76, 17525–17539 (2017). https://doi.org/10.1007/s11042-016-4121-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-016-4121-8