Machine Learning and Big Data for Cybersecurity: Systematic Literature Review

Today’s advances in wireless communication technology have resulted in the generation of large amounts of data. Networks face a large amount of data every moment. The utilization of advanced technologies, systems, and practices, intends to protect networks, organizations, projects, and data from assaults, attacks, harms, or unapproved access, which is defined as Cybersecurity. Cyberattacks have increased rapidly in various domains. For security reasons, cybersecurity solutions must search through ever-growing mountains of data for possible intrusion patterns. Therefore, it becomes necessary in such environment and conditions to detect intrusions in a fast and accurate way in order to protect data. The extent of this System Literature Review (SLR) includes an investigation of the most well-known artificial intelligence technologies applied to cybersecurity. This research work was conducted on multiple databases (Springer Link, SCOPUS, ScienceDirect, IEEE Xplore, ACM, and MDPI). From these sources, 35 articles published during the range of 2014 and 2023 were chosen, investigated and thoroughly assessed. This survey analyses Machine Learning and Big Data techniques used in cybersecurity and their adequacy in detecting attacks. Different Machine Learning and Deep Learning algorithms are presented and compared. Then, an in-depth analysis based on some metrics criteria: Accuracy, Precision, Detection Rate, False Positive Rate, Training Time and Prediction Time, is performed. Finally, the limitations encountered during this research work are presented along future work related to the studied topic.