A Semi-Supervised Sparsity Discriminant Analysis Algorithm for Feature Extraction

Abstract:

Recently, l₁-graph was proposed as a new graph construction procedure. Compared with the kNN-graph and ε-graph, l₁-graph possesses three advantages: robustness to data noise, sparsity and datum-adaptive neighborhood selection. In this paper, we propose a novel semi-supervised feature extraction method based on l₁-graph termed Semi-supervised Sparsity Discriminant Analysis (S³DA). The proposed S³DA maintains the advantages of l₁-graph, and more importantly, it has better capacity of discrimination for classification. Experimental results on face and gene expression databases demonstrate our proposed approach outperform some other state of the art algorithms, and also show the feasibility and effectiveness of our proposed approach.

• Periodical: Advanced Materials Research (Volumes 546-547)
• Pages: 670-674
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.546-547.670
• Online since: July 2012
• Authors: Hui Sun, Yu Gen Yi, Ying Hua Lv, Hui Tao Cai, Jian Zhong Wang
• Keywords: Face Recognition, Feature Extraction, l₁-Graph, Microarray Classification, S³DA

References

[1] J. Z. Wang, B. X. Zhang, M. Qi, J. Kong: Linear discriminant projection embedding based on patches alignment, Image and Vision Computing, 1624-1636 (2010).

DOI: 10.1016/j.imavis.2010.05.001

[2] I. T. Jolliffe, Principal Component Analysis, Second Ed., Springer-Verlag, New York, (2002).

[3] X.F. He, P. Niyogi: Locality preserving projections, In: Proceedings of the 17th annual conference on neural information processing systems, 153-160 (2003).

[4] K. Fukunaga: Introduction to Statistical Pattern Recognition, Second Ed., Academic Press, Boston (1990).

[5] D. Cai, X.F. He, J.W. Han: Semi-supervised discriminant analysis, in: Proceedings of the 11th IEEE International Conference on Computer Vision, ICCV07, Rio de Janeiro, Brazil, (2007).

DOI: 10.1109/iccv.2007.4408856

[6] Y.Q. Song, F.P. Nie, C.S. Zhang: Semi-supervised sub-manifold discriminant analysis, Pattern Recognition, 1806-1813 (2008).

DOI: 10.1016/j.patrec.2008.05.024

[7] M. Sugiyama, T. Id, S. Nakajima, J. Sese: Semi-supervised local Fisher discriminant analysis for dimensionality reduction, Machine Learning, 35-61 (2010).

DOI: 10.1007/s10994-009-5125-7

[8] R. Chatpatanasiri, B. Kijsirikul: A unified semi-supervised dimensionality reduction framework for manifold learning, Neurocomputing, 1631-1640 (2010).

DOI: 10.1016/j.neucom.2009.10.024

[9] B. Cheng, J. Yang, S. Yan, Y. Fu and T. Huang: Learning with l1-graph for image analysis, IEEE Transactions on Image Processing, 858-866 (2010).

DOI: 10.1109/tip.2009.2038764

[10] F. Zang, J.S. Zhang: Disciriminative learning by sparse representation for classification, Neurocomputing, 2176–2183 (2011).

DOI: 10.1016/j.neucom.2011.02.012

[11] A. Georghiades, P. Belhumeur, D. Kriegman: From few to many: Illumination cone models for face recognition under variable lighting and pose, IEEE Trans. Pattern Anal. Mach. Intell, 643-660 (2001).

DOI: 10.1109/34.927464

[12] F.S. Samaria, A.C. Harter: Parameterisation of a stochastic model for human face identification, in: IEEE Workshop on Applications of Computer Vision, 138-142 (1994).

DOI: 10.1109/acv.1994.341300

[13] U. Alon, N. Barkai, D.A. Notterman, K. Gish, S. Ybarra, D. Mack, A.J. Levine: Broad patterns of gene expression revealed by clustering analysis of tumor and normal colon tissues probed by oligonucleotide arrays, Proc. Natl. Acad. Sci. USA 6745-6750 (1999).

DOI: 10.1073/pnas.96.12.6745