Support vector machine model with discriminant graph regularization term

Traditional SVM classification model constructs linear discriminant function by maximizing the margin between two classes, and the weight vector of the discriminant function is only related to a small number of support vectors near the decision boundary. The small amount of support vectors is hard to describe the global distributive information when the distributions of the samples are nonlinear manifolds structure. To solve this problem, the graph regularization term with discrimination information is introduced into the objective function of SVM model. Experimental results on public data sets show that the classification accuracy of this method has improved significantly compared to traditional SVM models.

[1]  Shigeo Abe DrEng.  Pattern Classification , 2001, Springer London.

[2]  Vijander Singh,et al.  An efficient approach for face recognition based on common eigenvalues , 2014, Pattern Recognit..

[3]  Tao Xiong,et al.  A combined SVM and LDA approach for classification , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[4]  Xiaolong Teng,et al.  Face recognition using discriminant locality preserving projections , 2006, Image Vis. Comput..

[5]  Boubakeur Boufama,et al.  A novel SVM+NDA model for classification with an application to face recognition , 2012, Pattern Recognit..

[6]  Jiawei Han,et al.  Non-negative Matrix Factorization on Manifold , 2008, 2008 Eighth IEEE International Conference on Data Mining.

[7]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[8]  Lei Wang,et al.  Efficient Spectral Feature Selection with Minimum Redundancy , 2010, AAAI.

[9]  Hong Wang,et al.  Grouping Subspace Segmentation , 2017 .