Intrusion Detection Systems Based on Machine Learning Approaches: A Systematic Review

Abstract

The proliferation of Internet use poses certain security problems for networks. Intrusion detection (ID) in cybersecurity technology is to recognize unexpected entry to or assaults on secured networked computers. In the research, many machine learning (ML) and deep learning (DL) algorithms have been used to tackle intrusion detection systems (IDS). Nevertheless, few publications examine and explain the present state of employing ML approaches to tackle ID issues. This systematic review (SR) analyzes 11 papers published between 2016 and 2021 that focused on developing single, hybrid, and ensemble classifiers. Similar research is evaluated based on their classifier designs, the datasets they used, and their conceptual frameworks. Recent accomplishments and limits in developing IDS systems using ML are presented and analyzed. In addition, many prospective study possibilities are offered.