Classification of Malware in HTTPs Traffic Using Machine Learning Approach

Abstract

Cybersecurity and cyberwar have become crucial for a world backed by continuous development and expansion of digitalization. In the current digital era, malware has become a significant threat for internet users. Malware spreads faster and poses a big threat to cyber security. Hence, network security measures have an important role to play for neutralizing these cyber threats. In our research study, we collected some malicious and self-generated benign PCAP’s and then applied a Random Forest (RF) machine learning algorithm to build a traffic classifier. The proposed classifier classifies the HTTPs traffic as benign or malicious one. Experimental results exhibit the average accuracy of 90% and a false-positive rate of 0.030 for RF classifier.