基于卡方检验和LDOF算法的入侵检测技术研究

Research on Intrusion Detection Technology Based on Chi-square Test and LDOF Algorithm

为了解决入侵检测领域中数据维度较高问题,本文提出了一种方法,首次将卡方检验和LDOF算法结合在一起用于入侵检测领域。先利用卡方检验进行特征选择,然后计算每条数据对象的LDOF因子,最后检测出异常数据。最终的实验结果表明,LDOF算法和卡方检验算法相结合的入侵检测方法在处理网络高维数据时,不仅可以有效地检测出异常数据,并且有着更快的检测速度和较低的误检率。

In order to solve the problem of high data dimension in the field of intrusion detection,this paper proposes a method which combines chi-square test and LDOF algorithm.The chi-square test is used to select features after which the LDOF factor of each data object is calculated,and then the abnormal data is detected.Final experimental results show that this method can not only effectively detect abnormal data but also has faster detection speed and lower false detection rate when treating high-dimensional network data.