A Systematic Review of Artificial Intelligence and Machine Learning Techniques for Cyber Security

The use of technologies, procedures, and practices, designed to protect networks, programs, and data from attacks, damages, or unauthorized access, are called cyber security. Research community has applied different methods in the area of detection of cyber security attacks. However, literature lacks a systematic literature review (SLR) to summarize the use of Artificial Intelligence (AI) and Machine Learning (ML) methods, specifically the use of classifiers, in the detection of cyber security attacks. To cover the vacuum, this paper presents a systematic literature review of existing classification algorithms, applied to the area of detection of cyber security attacks. Relevant literature, qualifying specialized search criteria, is retrieved and extracted from the online libraries of Science Direct and Google Scholar. Total 63 research articles were found in these libraries, which are further filtered and refined to 21 articles. These 21 articles are critically reviewed and the following information are extracted: method used, classifier used, problem solved and domain selected. From the extracted information, a detailed taxonomy is prepared and presented to help beginners in the area to understand the problem. As a result of meta-level analysis, it is concluded that Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT) and Artificial Neural Network (ANN) are the most frequently used classifiers in the area of detection of cyber security.