PSO-Based Uncorrelated Hybrid Discriminant Analysis Algorithm

Abstract:

Hybrid discriminant analysis (HDA) can overcome small sample problems and outperform PCA and LDA by unifying principal component analysis (PCA) and linear discriminant analysis (LDA) in a single framework. However, the existing HDA algorithm can’t extract more discriminant information from dataset, and model parameters are difficult to select. To deal with the above problems, a particle swarm optimal (PSO)-based uncorrelated hybrid discriminant analysis algorithm is presented. The conjugate orthogonal condition is added to optimization problem of HDA, PSO is explored to select optimal HDA parameters and the optimal solution can be achieved by solving eigenvalue problem. Simulation demonstrates merits of the proposed algorithm.

• Periodical: Applied Mechanics and Materials (Volume 109)
• Pages: 671-675
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.109.671
• Online since: October 2011
• Authors: Xiao Ping Liu, Gui Yun Xu
• Keywords: Feature Extraction, HDA, Particle Swarm Optimization Algorithm (PSO), Statistical Uncorrelation

References

[1] W.H. Yang, D.Q. Dai, H. Yan, Feature extraction and uncorrelated discriminant analysis for high-dimensional data, IEEE Trans. Knowledge and Data Engineering. 8 (2007) 1-14.

DOI: 10.1109/tkde.2007.190720

[2] P. Howland, J . Wang, H. Park, Solving the Small Sample Size Problem in Face Recognition Using Generalized Discriminant Analysis, Pattern Recognition. 39 (2006) 277-287.

DOI: 10.1016/j.patcog.2005.06.013

[3] L. Wang, K. L. Chan, P. Xue, L. Zhou, A kernel-induced space selection approach to model selection in KLDA, IEEE Trans. Neural Networks. 12 ( 2008) 2116-2122.

DOI: 10.1109/tnn.2008.2005140

[4] W. S. Chen, P. C. Yuen, J. Huang, Kernel Machine based One- Parameter Regularized Fisher Discriminant Method for Face Recognition, IEEE Trans. Systems, Man, and Cybernetics-part B: Cybernetics. 4 (2005) 657-669.

DOI: 10.1109/tsmcb.2005.844596

[5] Q. Tian, J. Yu, T. S. Huang, Boosting Multiple Classifiers Constructed by Hybrid Discriminant Analysis, LNCS. 3541 (2005) 653-657.

[6] J. Yu, Q. Tian, T. Rui, Integrating discriminant and descriptive information for dimension reduction and classification, IEEE Trans. Circuits and Systems for Video Technology. 3 (2007) 372-377.

DOI: 10.1109/tcsvt.2007.890861

[7] A. K. Qin, P. N. Suganthan, M. Loog, Uncorrelated Heteroscedastic Lda Based on the Weighted Pairwise Chernoff Criterion, Pattern Recognition. 4 (2005) 613-616.

DOI: 10.1016/j.patcog.2004.09.007

[8] H. L. Xiong, M.N. S. Swanmy, M. Omair Ahmad, Optimizing the kernel in the empirical feature space, IEEE Trans. Neural Network. 3 (2005) 460-474.

DOI: 10.1109/tnn.2004.841784

[9] Q. L. Wang, X. Z. Gao, C. B. Wang, Swarm immune algorithm for PID controller self-tuning, Systems Engineering- Theory & Practice. 6 (2010) 1062-1066.