Self -Adaptive Parameter Optimization Approach for Least Squares Support Vector Machines

摘要：

Based on radial basis function (RBF) kernel, a new self-adaptive method to optimize the least squares support vector machines (LS-SVM) parameters, the width of kernel parameter σ and the LS-SVM regularization parameter γ are proposed. Detailed methodology steps of this algorithm method are presented. Compared with back propagation neural networks (BPNN), various simulation experiments for nonlinear function estimation are carried out. The results show that this prediction model can achieve higher identification precision with a reasonably small size of training sample sets and has high generalization performance.

关键词： Non-linear system Least squares support vector machines Error precision Prediction model

作者: Li Chun-xiang Zhang Wei-min Zhong Bi-liang

作者单位: Department of Computer Science and Information Technology, Guangzhou Maritime College, Guangzhou 510725, China

会议类型: 国际会议

会议名称: 2009年中国控制与决策会议(2009 Chinese Control and Decision Conference)

会议地点: 广西桂林

会议语种:英文

页码: 3516-3519

在线出版日期: 2009-06-17（万方平台首次上网日期，不代表论文的发表时间）