| 基于CPSO的混合核函数SVM参数优化及应用 |
| |
| 引用本文： | 潘丰,毛志亮.基于CPSO的混合核函数SVM参数优化及应用[J].控制工程,2011,18(2):267-269,274. |
| |
| 作者姓名： | 潘丰 毛志亮 |
| |
| 作者单位： | 江南大学,通信与控制工程学院,江苏,无锡,214122 |
| |
| 摘 要： | 支持向量机(SVM)建模的拟合精度和泛化能力取决于相关参数的选取,目前SVM中的参数的寻优一般只针对惩罚系数和核参数,而混合核函数的引入,使SVM增加了一个可调参数.针对混合核函数SVM的多参数选择问题,提出利用具有较强全局搜索能力的混沌粒子群(CPSO)优化算法对混合核函数SVM建模过程中的重要参数进行优化调整,每一...
|
| 关 键 词： | 建模 混合核函数 支持向量机 粒子群优化 |
Parameter Optimization and Application of SVM with mixture kernels Based on CPSO |
| |
| PAN Feng,MAO Zhi-liang.Parameter Optimization and Application of SVM with mixture kernels Based on CPSO[J].Control Engineering of China,2011,18(2):267-269,274. |
| |
| Authors: | PAN Feng MAO Zhi-liang |
| |
| Affiliation: | PAN Feng,MAO Zhi-liang(School of Communication and Control Engineering,Jiangnan University,Wuxi 214122,China) |
| |
| Abstract: | The parameters optimization problem to get the good regression accuracy and generalization performance of the support vector machine(SVM) is discussed.With the introduction of mixed kernels,SVM has one more adjustable parameter besides penalty and kernel parameter.To the multi-parameter selection problem of SVM with mixture kernels,a method is put forward by using chaotic particle swarm optimization which has better global search ability.Each particle indicates a group of the SVM parameters.The algorithm in... |
| |
| Keywords: | modeling mixture kernels support vector machine(SVM) particle swarm optimization(PSO) |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
