Multiclass Least Squares Complex Wavelet Support Vector Machines

摘要：

A complex wavelet support vector (SV) kernel,namely modified Morlet wavelet kernel,is proposed based on theoretic analysis.It is proved that this kernel function satisfies the Mercer.s condition and reproducing property in the Hilbert space.The reproducing property of the wavelet kernel function enhances the generalization ability of the support vector machines (SVM).Combining multiclass least squares support vector machine (MLS-SVM) with the complex wavelet kernel,a novel notion of multiclass least squares complex wavelet support vector machine (MLS-CWSVM) is proposed.The spiral multiclass classification experimental results show some advantages of MLS-CWSVM over MLS-SVM on the classification and the generalization performance.

关键词： complex wavelet Mercer condition Least squares support vector machine multiclass classification.

作者: Yong-zhong Xing Xiao-bei Wu Zhi-liang Xu

作者单位: Institute of Automation,Nanjing University of Science and Technology,Nanjing,210094 China

会议类型: 国际会议

会议名称: International Conference on Modelling,Identification and Control(模拟、鉴定、控制国际会议)

会议地点: 上海

会议语种:英文

在线出版日期: 2008-06-29（万方平台首次上网日期，不代表论文的发表时间）