Adaptive Machine Learning Strategies for Next-Generation Botnet Host Detection
Aniket Jhariya, Dhvani Parekh, Anurag Mogal, Joshua Lobo, Mangal Singh · 2025
The escalation in the number of Internet of Things (IoT) devices has led to a notable rise in the danger of botnet assaults, which pose a substantial risk to network security. This chapter uses deep learning (DL) and sophisticated machine learning approaches to address the important problem of identifying botnet hosts in IoT environments. To improve the accuracy of anomaly detection, we provide a novel detection paradigm that makes use of balanced datasets and dimensionality reduction. Our experimental results show an improvement in detection rates over current approaches, verified on several datasets. The results show that our method not only works better than conventional detection methods, but it can also be scaled and adjusted to meet the needs of changing threats. By providing a strong, this research advances the field of botnet identification and opens the door to more secure IoT networks. This chapter envisions a more secure digital landscape and the reconfiguration of cybersecurity paradigms, and it calls for an increase in the research community’s efforts, especially in leveraging emerging technologies like DL, anomaly detection techniques, and the integration of blockchain solutions. The study’s conclusion urges interdisciplinary cooperation, careful methodological improvements, and innovative culture with the shared goal of filling existing gaps and advancing botnet detection systems’ level of sophistication.