학술논문

Comparing and evaluating supervised and unsupervised machine learning techniques for IoT security.
Document Type
Article
Source
AIP Conference Proceedings. 2023, Vol. 2814 Issue 1, p1-9. 9p.
Subject
*SUPERVISED learning
*INTRUSION detection systems (Computer security)
*INTERNET of things
*TELECOMMUNICATION
*INDUSTRY 4.0
Language
ISSN
0094-243X
Abstract
The Internet of Things (IoT) technology expansion allows billions of globally distributed physical items to capture, collect, exchange, and share massive amounts of data. These physical entities include all connectable gadgets, from simple household devices to sophisticated industrial devices. Internationally, INDUSTRY 4.0 has altered the industrial sector due to the increasing availability and affordability of microprocessors and sensors, as well as the pervasiveness of communication technology. Connecting so many disparate things and implanting sensors in them gives gadgets a degree of digital intelligence, allowing them to interact instantaneously without human intervention. This paper provides an in-depth examination of supervised and unsupervised models for IoT security and intrusion detection. The model was used to classify solutions in a complete, organized evaluation of their applicability to IoT cybersecurity and their contribution to the development of efficient IoT intrusion detection systems. [ABSTRACT FROM AUTHOR]