摘要
支持向量机(SVM)是一种非常有前景的学习机器。然而在实际应用中,SVM的参数选取问题一直没有得到很好的解决,这在很大程度上限制了它的应用。为了能够自动地获得最佳参数,提出了基于遗传算法的SVM参数选取方法。该方法首先通过分析SVM参数对其性能的影响来确定遗传算法的搜索区间,然后在该区间内对搜索的参数进行选取。将该文提出的方法应用于5个由R tsch收集的标准模式库,实验结果表明由该方法所得参数确定的SVM具有较优的识别率和较简单的结构,即具有较佳的整体性能。
Support Vector Machines (SVM) is a promising learning technique. While in practice, the problem on how to select parameters of SVM is not solved properly. In order to get the optimal parameters automatically, a new approach based on genetic algorithm was proposed, which can acquire the best parameters of SVM. This method defines the search area by analyzing the behavior of SVMs with different parameters that also have different influence on the classrate and then chooses the best parameters in the given region. The method is experimented with five benchmark repotsch, the results demonstrate that the algorithm can get the SVM with the best recognition accuracy and simple structure.
出处
《辽宁石油化工大学学报》
CAS
2004年第1期54-58,共5页
Journal of Liaoning Petrochemical University
关键词
支持向量机
SVM
参数选取
遗传算法
统计学习理论
Statistic learning theory
Support vector machines
Genetic algorithm
Parameter selection
