Analysis of Intrusion Detection Systems in UNSW-NB15 and NSL-KDD Datasets with Machine Learning Algorithms

Abstract

The use of intelligent devices in almost every sector, and the provision of services by private and public institutions through network servers, cloud technologies, and database systems are now mostly remotely controlled. Due to the increasing demands on network systems, unfortunately, both malicious software and users are showing more interest in these areas. Some organizations are facing almost hundreds or even thousands of network attacks daily. Therefore, it is not enough to solve the attacks with a virus program or a firewall. Detection and accurate analysis of network attacks are crucial for the operation of the entire system. With the use of deep learning and machine learning, attack detection, and classification can be successfully performed. This study conducted a comprehensive attack detection process on the UNSW-NB15 and NSL-KDD datasets using existing machine learning and deep learning algorithms. In the UNSW-NB15 dataset, an accuracy of 98.6% and 98.3% was achieved for two-class and multi-class classification, respectively, and 97.8% and 93.4% accuracy were obtained in the NSL-KDD dataset. The results prove that machine learning algorithms are an effective solution for intrusion detection systems.