APT Guard- Preventive Measures Against APT Attack

Abstract

Many researchers have put their efforts into defence against this Advanced Persistence Threat (APT) attack. The traditional security systems such as web and email protectors and canners are no longer suitable for defending and preventing damages. The proposed system helps to detect ATP attacks from the network traffic data using a convolution Neural network (CNN). The experimentation is performed on the NSL-KDD dataset. Feature engineering is a major part of the system where we can select the 14 most appropriate features among 42 available features. One of the most effective approaches to APT attack detection is to use machine learning to analyse network traffic. The proposed method gives superior results in the detection of APT attacks.