Toward the Use of Machine Learning and Ensemble Learning Algorithms for IDS in the IoT Domain

The use of intrusion detection systems is one of the means to provide a secure and reassuring environment for computer system users. Updates and improvements for these protection systems are recommended due to the regular appearance of new vulnerabilities. Research has demonstrated the significant role of Machine Learning in the conception and implementation satisfactory techniques that can predict new attacks more quickly and efficiently to prepare the most suitable countermeasure. The main concern of this article is to establish a comparative study and implementation of Machine Learning and Ensemble Learning algorithms for IDS in the IoT domain. More specifically, the focus is directed towards the study and implementation of IDS using datasets containing real data, and through the utilization of unitary and ensemble learning algorithms, effective solutions can be achieved to enhance the detection capacity of intrusion detection systems, Furthermore, another part of the work involves initiating and setting up a Snort IDS on a Raspberry Pi board, thus adding a practical perspective to our paper.