Abstract: Smoothing functions play an important role in Support Vector Machines（SVMs）.Using the technique of interpolation functions，Xiong et al developed a recursive formula to obtain a new class of smoothing functions.Their work is a major step to the problems of existence and seeking better smooth functions.However，it remains unclear the performance of the 5th-order smooth functions. To that end，this paper first derived the expression of such smooth functions.Then it offered a detailed analysis of its properties.The results showed that its approximation accuracy is competitive compared to previously used smooth functions.

Key words: pattern recognition, classification, support vector machine, smoothing

摘要： 光滑函数在支持向量机中起着重要作用，熊金志等人用插值函数的方法导出了一个递推公式，得到了一类新的光滑函数，解决了一个关于是否存在以及如何寻求性能更好的光滑函数的问题。然而其中五阶光滑的多项式函数还未进行性能分析。首先推导出了该光滑函数的表达式，然后分析了它的若干性能。结果表明，该光滑函数具有良好的逼近性能，其逼近精度高于以往的光滑函数。

关键词: 模式识别, 分类, 支持向量机, 光滑