Abstract: To improve classific precision of network intrusion detection model and reduce the number of training data set and learning time, a new supervisal algorithm based on ε Support Vector Regression machines (ε-SVR) machine was proposed. Firstly, normalization was used on training data set, and then a new coefficience of sparse penalty function was adjusted. Finally, the experimental results using KDD CUP 1999 data set show that this approach can detect intrusion behavior, increase its veracity and validity, and reduce its distortion.

Key words: network security, intrusion detection, ε Support Vector Regression （ε-SVR） machine, algorithm design

摘要： 为提高网络入侵检测系统中检测算法的分类精度，降低训练样本及学习时间，提出一种新的基于支持向量回归机的检测算法。算法首先归一化处理训练样本数据，然后精确调节松弛惩罚因子，最后使用KDD CUP 1999数据集进行仿真实验，结果表明本算法可以提高入侵检测的准确性和有效性，并能够降低误报率。

关键词: 网络安全, 入侵检测, ε支持向量回归机, 算法设计