Network Intrusion detection by using Feature Reduction Technique

摘要: As network attacks have increased in number and severity over the past few years, intrusion detection system (IDS) is increasingly becoming a critical component to secure network. Due large volumes of security audit data as well complex dynamic properties behaviors, optimizing performance IDS becomes an important open problem that receiving more attention from research community. Intrusion poses serious risk environment. The ever growing new types pose for their detection. In this paper, method based on Principle Component Analysis (PCA) Random Forest with low overhead high efficiency presented. System call command sequences are used information sources validate proposed method. frequencies individual calls trace commands block computed then column vectors which represent traces blocks formed input. PCA applied reduce dimensional distance between vector its projection onto subspace reduced anomaly Experimental results show promising terms accuracy, computational expense implementation real-time