数据挖掘技术在入侵检测中的应用与研究

决策树是数据挖掘的分类应用中采用最广泛的模型之一,但是传统的ID3、C4.5和CART等算法在应用于超大型数据库的挖掘时,有效性会降得很低,甚至出现内存溢出的现象,针对此本文提出了一种基于属性加权的随机决策树算法,并通过实验证明该算法减少了对系统资源的占用,并且对高维的大数据集具有很高的分类准确率,非常适合被用于入侵检测的分类之中。

Application and research of data mining technology on intrusion detection system

Decision tree is one of the most widely using model in classification of the data mining,but when the traditional decision tree algorithms,such as ID3,C4.5 and CART,are applied in large database,the effectiveness is lower,and even apear memory overflow.So this article proposes a new algorithm that is random decision tree algorithm.This algorithm is proved that it reduces occupied of the system resources,and has a higher classification accuracy in large data sets,so be very suitable for the classification of intrusion detection.