摘要
核函数是SVM的关键技术,核函数的选择将影响着支持向量机的学习能力和泛化能力。各个普通核函数各有利弊,在分析各个普通核函数的基础上,采用了一种新的组合核函数,它既具有很好的泛化能力,也具有很好的学习能力,并将其构造的支持向量机应用到网络安全的风险评估中,与普通核函数构造的支持向量机的评估效果进行比较。结果表明组合核函数支持向量机不仅提高了分类速度,而且具有较高的分类精度。
Kernel function is the key technology of SVM,the choice of kernel will affect the learning ability and generalization ability of SVM.Since every traditional kernel has its advantages and disadvantages,this paper analyzes the principle of traditional kernels and adopts a new kernel of combined kernel which has better generalization ability and better learning ability,and adopts the combined kernel SVM into network security risk evaluation,and then compares with the SVM using traditional kernels.The results show that the SVM based on combined kemels has batter speed and higher force of classification than that with traditional kernels.
出处
《计算机工程与应用》
CSCD
北大核心
2009年第11期123-125,共3页
Computer Engineering and Applications
关键词
支持向量机
组合核函数
网络安全
风险评估
Support Vector Machine（SVM）
combined kernel
network security
risk evaluation
