计算机科学 ›› 2015, Vol. 42 ›› Issue (6): 239-242.doi: 10.11896/j.issn.1002-137X.2015.06.050
田浩兵,朱嘉钢,陆 晓
TIAN Hao-bing, ZHU Jia-gang and LU Xiao
摘要: 粗糙one-class支持向量机(ROCSVM)是一种一类支持向量机,它通过核函数映射,定义上近似超平面和下近似超平面,使得训练样本能根据在粗糙间隔中的位置,自适应地对决策超平面产生影响。由于ROCSVM训练集只有正类样本,因此充分挖掘和利用训练样本的分类特征对于提高ROCSVM的分类性能有重要意义。为此,提出了一种基于训练样本分类特征贡献度的加权高斯核函数(λ-RBF):先对训练样本做主成分分析(PCA)得到按特征值排序的向量集,以此向量集构造核函数,使得特征值较大的维度在核函数中起较大的作用。在UCI标准数据集和仿真数据上的实验结果表明:与一般RBF的ROCSVM相比,基于λ-RBF的ROCSVM有着更好的泛化性和更高的识别率。
[1] Tax D M J,Duin R P W.Support Vector Data Description [J].Machine Learning,2004,54(1):45-66 [2] Campbell C,Bennett P.A Linear Programming Approach toNovelty Detection [M].Advances in Neural Information Processing Systems.Cambridge:MIT Press,2001 [3] Wang L,Jia H D,Li J.Training Robust Support Vector Machine with Smooth Ramp Loss in the Primal [J].Neurocomputing,2008,71:3020-3025 [4] Lin C F,Wan g S D.Fuzzy Support Vector Machine [J].IEEE Transactions on Neural Networks,2002,13(2):464-471 [5] Xu Yi-tian,Liu Chun-mei.A rough margin-based one class support vector machine[J].Neural Comput & Applic,2013,22:1077-1084 [6] Bishop C M.Pattern Recognition and Machine Learning [M].Cambridge:Springer,2007:291-320 [7] 王磊,杨一帆,周启海.粗糙one-class支持向量机[J].计算机科学,2009,6(9):242-245 Wang Lei,Yang Yi-fan,Zhou Qi-hai.Rough Set based One-class Support Vector Machine[J].Computer Science,2009,6(9):242-245 [8] 秦玉平,王袆,伦淑娴,等.基于超椭球支持向量机的兼类文本分类算法[J].计算机科学,2013,0(11A):98-100 Qin Yu-ping,Wang Yi,Lun Shu-xian,et al.Multi-label Text Classification Algorithm Based on Hyper Ellipsoidal SVM[J].Computer Science,2013,0(11A):98-100 |
No related articles found! |
|