计算机科学 ›› 2014, Vol. 41 ›› Issue (5): 245-249.doi: 10.11896/j.issn.1002-137X.2014.05.052
胡文军,王娟,王培良,王士同
HU Wen-jun,WANG Juan,WANG Pei-liang and WANG Shi-tong
摘要: 线性SVM具有算法简单、训练和测试速度快等优点,但不能用于解决线性不可分问题。为此,将样本数据集划分为多个集合并分别构造它们的LSVM,然后运用径向基函数的非线性组合来拟合非线性的决策函数,从而解决线性不可分问题。鉴于此,提出了一种适合非线性大样本分类的LSVM快速集成模型FMELSVM。该模型利用径向基函数RBF改善了LSVM的非线性输出能力,同时引进了优化权来提升LSVM的集成效果。UCI数据集的实验结果表明,FMELSVM在处理大样本方面具有较好的性能优势。
[1] Cortes C,Vapnik V.Support vector networks[J].Machine Learning,1995,20(3):273-297 [2] Schlkopf B,Smola A,Williamson R C,et al.New support vector algorithms[J].Neural Computation,2000,12(5):1207-1245 [3] 王晓明,王士同.平均邻近间隔支撑向量机[J].智能系统学报,2010,5(4):313-319 [4] 皋军,王士同,邓赵红.基于全局和局部保持的半监督支持向量机[J].电子学报,2010,38(7):1626-1633 [5] Tsang I W,Kwok J T,Cheung P M.Core vector machines:Fast SVM training on very large data sets[J].Journal of Machine Learning Research,2005,6:363-392 [6] Deng Zhao-hong,Chung Fu-lai,Wang Shi-tong.FRSDE:fast reduced set density estimator using minimal enclosing ball appro-ximation[J].Pattern Recognition,2008,41(4):1363-1372 [7] Chung Fu-lai,Deng Zhao-hong,Wang Shi-tong.From minimum enclosing ball to fast fuzzy inference system training on large datasets[J].IEEE Transactions on Fuzzy Systems,2009,17(1):173-184 [8] Tran Q A,Zhang Q L,Li X.Reduce the number of support vectors by using clustering techniques[C]∥International Confe-rence on Machine Learning and Cybernetics.Xi’an,China,2003:1245-1248 [9] JooSeuk K,Clayton D S.L2 kernel classification[J].IEEETransactions on Pattern Analysis and Machine Intelligence,2010,32(10):1822-1831 [10] Platt J C.Fast training of support vector machines using sequential minimal optimization[M]∥Schlkopf B,Burges C J C,Smola A J.Advances in kernel methods. Cambridge,MA: MIT Press,USA,1999:185-208 [11] Thorsten J.Training linear SVMs in linear time[C]∥Proc 12th ACM International Conference Knowledge Discovery and Data Mining.Philadelphia,PA,2006:217-226 [12] Hsieh C J,Chang K W,Lin C J,et al.A dual coordinate descent method for large-scale linear SVM[C]∥Proc 25th International Conference Machine Learning.Helsinki,Finland,2008:1-12 [13] Fan R E,Chang K W,Hsieh C J,et al.LIBLINEAR:A library for large linear classification[J].Journal of Machine Learning Research,2008,9:1871-1874 [14] 胡文军,王士同,王娟,等.非线性分类的分割超平面快速集成方法[J].电子信息学报,2012,18(7):843-860 [15] Platt J C.Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods[M]∥Smola A J,Bartlett P,Schlkopf B,et al.Advances in Large Margin Classifiers.Cambridge:MIT Press,1999:61-74 [16] Stefan R.SVM classifier estimation from group probabilities[C]∥Proc 27th International Conference Machine Learning.Haifa,Israel,2010:911-918 [17] Kovalsky S Z,Cohen G,Hagege R,et al.Decoupled linear estimation of affine geometric deformations and nonlinear intensity transformations of images[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(5):940-946 [18] Banach S.Sur les opérations dans les ensembles abstraits et leur application aux équations intégrales[J].Fundamenta Mathema-ticae,1922,3:133-181 [19] Tao Jian-wen,Wang Shi-tong,Hu Wen-jun,et al.ρ-Margin kernel learning machine with magnetic field effect for both binary classification and novelty detection[J].International Journal of Software and Informatics,2010,4(3):305-324 |
No related articles found! |
|