ML-Driven Lightweight Botnet Detection System for IoT-Networks
Journal Title: International Journal of Innovations in Science and Technology - Year 2024, Vol 6, Issue 7
Abstract
The integration of cloud computing with the Internet of Things (IoT) seeks to create seamless connections between humans and devices, enhancing applications in areas like smart healthcare and home automation. However, this also brings significant security challenges. Our study addresses the critical need for an efficient anomaly detection system specifically designed for IoT-enabled cloud computing environments, a gap not previously explored at this scale. Utilizing the IoT-23 dataset, we evaluated various feature selection techniques in conjunction with classification algorithms to develop a lightweight anomaly detection model. Our results demonstrate that the decision tree classifier, paired with the correlation coefficient method for feature selection, achieved an impressive 99.98% accuracy rate, with an average processing time of just 5.2 seconds. This combination proved to be the most effective for real-time anomaly detection, presenting a promising approach for ensuring robust security in IoT networks as connectivity continues to grow.
Authors and Affiliations
Ashfaq Hussain Farooqi, Rizwan Ahmad, Shaharyar Kamal
Keywords
Related Articles
- EP ID EP761762
- DOI -
- Views 28
- Downloads 0
LULC-NEAT: Land Use Land Cover Classification Using NeuroEvolution of Augmenting Topologies
Introduction/Importance of Study: NEAT's potency in optimizing neural networks for accurate LULC classification, aimed at better environmental stewardship, is shown. Novelty statement: LULC-NEAT introduces NeuroEvoluti...
Enhancing Face Mask Detection in Public Places with Improved Yolov4 Model for Covid-19 Transmission Reduction
Over the past decade, computer vision has emerged as a pivotal field, focusing on automating systems through the interpretation of images and video frames. In response to the global impact of the COVID-19 pandemic, the...
Breaking Down Monoliths: A Graph Based Approach to Microservices Migration
Introduction: The software industry has increasingly transitioned from Monolithic Architecture (MA) to Microservices Architecture (MSA) due to the significant advantages offered by MSA. A crucial first step in this mig...
Pedagogical Suitability: A Software Metrics-Based Analysis of Java and Python
Programming is one of the foundational skills essential for computer science professionals, yet attaining proficiency in this skill is widely acknowledged as a formidable challenge. The intrinsic complexity of program...
Enabling Early Treatment: A Deep Learning Approach to Multi-Class Potato Leaf Disease Identification
Over 60% of the world's population largely depends on the agricultural sector for food, as indicated by previous studies, demonstrating the historical significance of agriculture as a means of survival. Plant infection...