基于TCM-KNN算法的数据预处理问题研究

A Study on the TCM-KNN Algorithm Based Data Preprocessing Problem

就基于数据预处理的入侵检测系统进行了研究,并针对数据预处理子系统,提出了一种有效的预处理方法,即将对数据源的基本处理和基于TCM-KNN算法的数据预处理聚类器相结合。实验证明,经过预处理的数据,不仅使不完整信息数量和攻击数据数量大大减少,而且使入侵检测系统的检测率得到了进一步的提高,误报率得到了进一步的降低。

This paper discussed the problem of data preprocessing of the invasion detection system. We investigated a data preprocessing based invation deteetion system in view of the problems of low check rate and high false alarm rate of the present system, and proposed an effective preprocessing approach to the data preprocessing subsystem, which integrated the basic processing for a data source and TCM-KNN algorithm based data preproeessing cluster. The experiment proves that the approach not only greatly decreases the imperfect information and attack data quantity, but also farther increases the detection rote and reduces the false alarm rate.