A Comparative Study between Deep Learning Algorithm and Bayesian Network on Advanced Persistent Threat(APT) Attack Detection.
Keywords:accuracy, APT attack detection, Bayesian Network, deep learning algorithm, NSL-KDD dataset
Advanced Persistent Threat(APT) attacks are a major concern for the cybersecurity in digital world due to their advanced nature. Attackers are skillful to cause maximal destruction for targeted cyber environment. These APT attack are also well funded by governments in many cases. The APT attacker can achieve his hostile goals by obtaining information and gaining financial benefits regarding the infrastructure of a network. It is highly important to study proper countermeasures to detect these attacks as early as possible due to sophisticated methods. It is difficult to detect this type of attack due to the fact that the network may crash because of high traffic. Hence, in this study, Bayesian network and deep learning algorithm are used for timely detection and classification of APT-attacks on the NSL-KDD dataset. Moreover, 10-fold cross validation method is used to experiment these models. Other criterion such as accuracy, sensitivity, ROC-curve and false negative rate are also compared for the models.