期刊名称：International Journal of Computers and Communications
印刷版ISSN：2074-1294
出版年度：2014
卷号：8
页码：134-139
出版社：University Press
摘要：Network traffic is increasing due to the growing use of smart devices and the Internet. Most intrusion detection studies have focused on feature selection or reduction because some features are irrelevant or redundant which results in a lengthy detection process and degrades the performance of an intrusion identify important selected input features for building an Intrusion Detection System (IDS) that is computationally efficient and effective. To this end, we investigated the performance of standard feature selection methods; CFS(Correlation-based Feature Selection), IG(Information Gain) and GR(Gain Ratio). In this paper, we propose a new feature selection method using feature average of total and each class and applied efficient classifier decision tree algorithm for evaluating feature reduction method. Moreover, we compared the proposed method and other methods.
