Network intrusion detection system: A systematic study of machine learning and deep learning approaches

• Main Authors: Zeeshan, Ahmad, Adnan Shahid, Khan, Cheah Wai, Shiang, Johari, Abdullah, Farhan, Ahmad
• Format: Article
• Language: English
• Published: Wiley 2021
• Subjects: QA76 Computer software
• Online Access: http://ir.unimas.my/id/eprint/37907/1/machine%20learning1.pdf; http://ir.unimas.my/id/eprint/37907/; https://onlinelibrary.wiley.com/toc/21613915/2021/32/1; https://doi.org/10.1002/ett.4150
• Institution: Universiti Malaysia Sarawak

The rapid advances in the internet and communication fields have resulted in ahuge increase in the network size and the corresponding data. As a result, manynovel attacks are being generated and have posed challenges for network secu-rity to accurately detect intrusions. Furthermore, the presence of the intruderswiththeaimtolaunchvariousattackswithinthenetworkcannotbeignored.Anintrusion detection system (IDS) is one such tool that prevents the network frompossible intrusions by inspecting the network traffic, to ensure its confidential-ity, integrity, and availability. Despite enormous efforts by the researchers, IDSstillfaceschallengesinimprovingdetectionaccuracywhilereducingfalsealarmrates and in detecting novel intrusions. Recently, machine learning (ML) anddeep learning (DL)-based IDS systems are being deployed as potential solutionsto detect intrusions across the network in an efficient manner. This article firstclarifiestheconceptofIDSandthenprovidesthetaxonomybasedonthenotableML and DL techniques adopted in designing network-based IDS (NIDS) sys-tems. A comprehensive review of the recent NIDS-based articles is provided bydiscussing the strengths and limitations of the proposed solutions. Then, recenttrends and advancements of ML and DL-based NIDS are provided in terms ofthe proposed methodology, evaluation metrics, and dataset selection. Using theshortcomings of the proposed methods, we highlighted various research chal-lenges and provided the future scope for the research in improving ML andDL-based NIDS