Abstract—The number of internet-connected devices is rising, resulting in a global network of connected devices, referred to as Internet of Things (IoT). The technologically advanced agriculture industry employs IoT to monitor their environment and automate required functionality. IoT devices generate enormous amount of confidential and critical data, hence, securing the information is of significant importance. This research proposes integrating computationally intensive Machine Learning (ML) classifiers with resource-constrained IoT devices in order to safeguard the obtained data. This study analyses Naïve Bayes and Decision Tree for a cutting-edge Edge-IIoTset cybersecurity dataset encompassing 15 classes of IoT traffic derived from Soil Moisture, Temperature, Humidity, Water Level, and Water pH Sensors to enhance IoT data security. The experimental results of both the ML classifiers on given subsets of Edge-IIoTset presented Decision Tree as superior option, achieving accuracy of 72% and 73% for ML and DNN Edge-IIoTset respectively as compared to Naïve Bayes with accuracy of 47% and 45% respectively.
 
Keywords—internet of things, anomaly detection, malicious activity classification, naïve bayes, decision tree


Copyright © 2023 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.