论文部分内容阅读
支持向量机表现的好坏很大程度上取决于核函数的选取,因此最近几年关于核函数的研究有许多。越来越多的核函数也被提了出来!但是选取合适的核函数往往却不容易,因为数据的特征往往不知道。文中利用函数的Taylor展开思想,提出了一种新的核函数,叫T-KMOD,基于KMOD提出的。该核函数的灵活性更好,可以处理很多分类的问题。用网络入侵的数据对该核函数进行了仿真,从仿真的结果可以看出,和一些常用的核函数相比,它的鲁棒性更好,有更强的分类能力。同时该函数的分类效果更好。所以该核函数和一般常用的核函数相比,可能更具有一般选择性。
SVM performance depends largely on the selection of kernel function, so in recent years there are many studies on the kernel function. More and more kernel functions have also been proposed! However, it is often not easy to select suitable kernel functions because the characteristics of the data are often not known. In this paper, we use the Taylor expansion of the function to propose a new kernel function called T-KMOD, which is based on KMOD. The kernel is more flexible and can handle many taxonomic issues. The kernel function is simulated by network intrusion data. It can be seen from the simulation results that it has better robustness and stronger classification ability than some commonly used kernel functions. At the same time the classification of the function better. Therefore, the kernel function and the more commonly used than the kernel function may be more general selectivity.