Published Paper


Approaches for IoT Cyber security Analysis Using Various Machine Learning

R. Karthigaichelvi, Dr. B. Balakumar
Centre for Information Technology and Engineering, Manonmanium Sundaranar University
Page: 629-635
Published on: 2023 March

Abstract

As a result of recent scientific breakthroughs, new technologies are often developed and presented. Since managing and overseeing systems like these can be challenging for humans, our society has turned to machine learning for assistance. New ideas and methods are brought about by new technologies, and new techniques are used to circumvent existing cybersecurity measures. Three alternative Internet of Things (IoT) cyber security algorithms are currently in use in business for malware and intrusion detection: K-Nearest Neighbour (KNN), Support-Vector Machine (SVM), and Random Forest (RF)and. Training and testing were conducted on each algorithm. For malware detection, the highest accuracy of the KNN, SVM, and RF was 90.81%, 84.51%, and 93.37%. For intrusion detection, the highest accuracy was 92.58%, 87.33%, and 93.97%.

PDF