Abstract The selection of kernel function and its parameters determine the performance of kernel function. A semi-supervised data-dependent kernel optimization algorithm is presented, which uses unlabeled data and pairwise constraints to maximize an objective function sensitive to data-dependent kernel, so that its performance is improved. Then the proposed method is employed to optimize the kernel of kernel principal components analysis (KPCA) and the experimental results of the classification and clustering performance on the artificial data and UCI data sets show its efficiency.
[1] Wang Na, Li Xia. Kernel Parameters Optimization for Semi-Supervised Fuzzy Clustering with Pairwise Constraints. Chinese Journal of Electronics, 2008, 17(2): 297-300 [2] Browne M W. Cross-Validation Methods. Journal of Mathematical Psychology, 2000, 44(1):108-132 [3] Huang C L, Wang C J. A GA-Based Feature Selection and Parameters Optimization for Support Vector Machines. Expert Systems with Applications, 2006, 31(2): 231-240 [4] Amari S, Wu S. Improving Support Vector Machine Classifiers by Modifying Kernel Functions. Neural Networks, 1999, 12(6): 783-789 [5] Xiong Huilin, Swamy M N, Ahmad M O. Optimizing the Kernel in the Empirical Feature Space. IEEE Trans on Neural Networks, 2005, 16(2): 460-474 [6] Li Junbao, Pan Jengshyang, Lu Zheming. Kernel Optimization-Based Discriminant Analysis for Face Recognition. Neural Computing and Application, 2009, 18(6): 603-612 [7] Chen Bo, Liu Hongwei, Bao Zheng. Optimizing the Data-Dependent Kernel under a Unified Kernel Optimization Framework. Pattern Recognition, 2008, 41(6): 2107-2119 [8] Chen Bo, Liu Hongwei, Bao Zheng, et al. A Kernel Optimization Algorithm Based on Fusion Kernel for High-Resolution Range Profiles Recognition. Acta Electronica Sinica, 2006, 34(6): 1146-1151 (in Chinese) (陈 渤,刘宏伟,保 铮,等.一种针对雷达高分辨距离像识别的融合核优化算法.电子学报, 2006, 34(6): 1146-1151) [9] Shao Jidong, Rong Gang, Lee J M. Learning a Data-Dependent Kernel Function for KPCA-Based Nonlinear Process Monitoring. Chemical Engineering Research and Design, 2009, 87(11): 1471-1480 [10] Wang Na, Li Xia, Cui Yingjie, et al. Instance-Level Based Discriminative Semi-Supervised Dimensionality Reduction with Chunklets. International Journal of Innovative Computing, Information and Control, 2010, 6(8): 3763-3774 [11] Zhang Daoqiang, Chen Songcan, Zhou Zhihua. Learning the Kernel Parameters in Kernel Minimum Distance Classifier. Pattern Recognition, 2006, 39(1): 133-135 [12] Mouchaweh M S. Semi-Supervised Classification Method for Dynamic Applications. Fuzzy Sets and Systems, 2010, 161(4): 544-563 [13] Wang Na, Li Xia. Active Semi-Supervised Spectral Clustering Based on Pairwise Constraints. Acta Electronica Sinica, 2010, 38(1): 172-176 (in Chinese) (王 娜,李 霞.基于监督信息特性的主动半监督谱聚类算法.电子学报, 2010, 38(1): 172-176)
2011, Vol. 24 
