Development Machine Learning Techniques to Enhance Cyber Security Algorithms.

Mohamed Amer, Ghada; Abd El Hay, Ehab Hany; Abdel-Baset, Ibrahim Yasser; Abd El Azim Mohamed, Mohamed

doi:10.21608/bfemu.2021.206401

Development Machine Learning Techniques to Enhance Cyber Security Algorithms.

Document Type : Research Studies

Authors

¹ Communications Engineer at North Delta Electricity Distribution Company, Faculty of Engineering, Mansoura University.

² Assistant Prof., Faculty of Engineering, Mansoura University.

³ Assistant Lecturer, Nile Higher Institute for Engineering and Technology. Faculty of Engineering, Mansoura University.

⁴ chief, Dean of the College and Chairman of the Board of Directors, Faculty of Engineering, Mansoura University.

10.21608/bfemu.2021.206401

Abstract

Nowadays, Cyber security threats are a growing global problem. As technology evolves, cyber threats, including cyber-hacking threats, and cybercrime organizing groups, are on the rise. Distributed Denial of Service (DDoS) is one of the most serious attacks faced by Cloud computing. This attack aims to make cloud services unavailable to end-users by exhausting system resources, resulting in heavy losses that pose a threat to national security and information security assets, and thus making the development of defensive solutions against such attacks necessary to expand the use of Cloud computing technology. Machine learning (ML) has promising results in detecting cyber-attacks including DDoS when applied to intrusion detection systems. In this research, the proposed system was built using Random forest (RF) is supervised machine learning algorithm, which is an ensemble learning method that operates by constructing a multitude of decision trees at training time. The experiments conducted using the most common and standard data sets, NSL-KDD, and CICIDS 2017, achieved a detection accuracy of up to 99.09% for the first dataset and 99.97% for the second dataset respectively. The proposed system performs well when compared to other methods in terms of accuracy, detection rate, and low false-positive rate.

Keywords