IoT Botnet Detection and Classification usingMachine Learning Algorithms

Abstract

This scholarly research paper addresses the crucial and complex challenge of detecting and categorizing Internet of Things (IoT) botnets through the utilization of machine learning algorithms. The study is focused on conducting meticulous analysis and manipulation of IoT botnet data, with a specific emphasis placed on the widely acknowledged IoT-23 dataset. The principal aim is to employ widely recognized and widely-used machine learning algorithms, encompassing Decision Trees (DT), k-Nearest Neighbors (KNN), Random Forests (RF), and eXtreme Gradient Boosting (XGBoost), with the purpose of effectively classifying and detecting botnets within the confines of the IoT-23 dataset. By implementing these algorithms, the paper seeks to augment our understanding of their performance and efficacy within the domain of IoT botnet detection and classification. The execution of a comparative analysis, contrasting the outcomes derived from the diverse algorithms, will furnish invaluable insights into their respective merits and constraints, thereby enabling researchers and practitioners to make informed decisions concerning the most suitable algorithm for achieving successful IoT botnet detection and classification