A Machine Learning Approach for Anomaly Detection of IoT Traffic
Pranay Chouksey
PG Scholar, Gyan Ganga Institute of Technology and Sciences, Jabalpur
Dr. Siddharth Bhalerao
Associate Professor, Gyan Ganga Institute of Technology and Sciences, Jabalpur
Abstract
The implementation of IoT technology has enhanced the quality of life by optimizing numerous practical applications. The Internet of Things encompasses several devices that produce substantial volumes of data, necessitating computationally demanding processing.Identifying vulnerabilities and ensuring security are primary priorities in the field of IoT. Network vulnerability detection is a highly promising and efficient method for enhancing network security. The machine learning model is trained to proficiently analyze traffic data. This document presents a comprehensive analysis of machine learning techniques used to identify vulnerabilities in Internet of Things (IoT) applications. This article discusses the primary concerns and obstacles involved in acquiring knowledge about the technology utilized for practical issues in the Internet of Things. The suggested solution utilises authentic IoT data derived from actual IoT traffic. The paper utilises many classification algorithms such as Regression, Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), k- Nearest Neighbor (KNN), and Multi-layer Perception (MLP). Results were achieved in traffic models that involved both dual and multi-class scenarios.The findings were evaluated based on different parameters, and it was determined that the proposed strategy is more effective.Keywords: IoT network attack detection, machine learning, multiclass classification, anomaly detection, Distributed Denial-of-Service (DDoS), Denial-of-Service (DoS), IoT Security
Keywords: IoT network attack detection, machine learning, anomaly detection, Distributed Denial-of-Service (DDoS), Denial-of-Service(DoS),IoT Security
