Classification of Malicious Traffic Using TensorFlow Machine Learning

Abstract

With the rapid development of the Internet and the innovative attacks, information security has become an important issue for system administrators and users. Because the traditional intrusion detection system is based on misuse detection technology, the disadvantage is that it needs constant updating of the feature database to cope with attacks from variant malware. This paper proposes a framework of deep learning model by using the TensorFlow platform and utilizes the NSL-KDD data set for training and testing the proposed framework. Experimental results show the proposed methodology can effectively classify malicious traffic categories.