Parameters Optimization of Support Vector Regression based on Cultural Algorithm

摘要：

Parameters are important factors to affect the performance of SVR. Parameters optimization of SVR based on cultural algorithm is presented to improve prediction accuracy and generalization ability. Optimal rule is the least mean square error of samples, and the parameters of SVR are optimized based on adaptive capacity of cultural algorithm. The simulation results show that the algorithm can accelerate the pace of the search parameters and improve prediction accuracy of SVR, It has good robustness and strong global search capability.

关键词： support vector regression (SVR) cultural Algorithm (CA) parameters optimization

作者: Kaifan Yang

作者单位: Apartment of Mathematics,Shaanxi University of Technology,Hanzhong, Shaanxi ,China

会议类型: 国际会议

会议名称: 2010 International Conference on Circuit and Signal Processing(2010年电路与信号处理国际会议 ICCSP 2010)

会议地点: 上海

会议语种:英文

页码: 215-218

在线出版日期: 2010-12-25（万方平台首次上网日期，不代表论文的发表时间）