局部影响分析在支持向量机中的应用

Application of local influence in support vector machines

摘要: 讨论了通过对支持向量的局部扰动来确定支持向量集中的强影响点,并由这些强影响点构造相同精度下具有更强广义能力的支持向量机;对线性核函数和高斯径向基函数所构建的SVM进行了分析,并给出一个实例.

Abstract: The recognition of strong influence points in support vector set by local perturbations is studied,since SVM with better generalized ability can be constructed by these strong influence points;for illustration,SVM based on linear kernel function and Radial Basis Function are analyzed respectively.Also a real example is given.