Malware Attack Detection on IoT Network using Machine Learning Techniques

Abstract

The use of Internet of Things (IoT) devices is growing rapidly in parallel with the expansion of the internet. As the data capacity of IoT devices grows, they are becoming more susceptible to malware assaults. Consequently, detecting malware on IoT devices has become a crucial concern. A technique that is both effective and dependable, while also saving time, is necessary for identifying complex malware. In recent years, researchers have put forth many techniques for identifying malware. Nevertheless, achieving precise detection still poses a significant hurdle. This research introduces a Hyperparameter Fine-tuned SVM (HFSVM) machine learning method for identifying malware on IoT devices. The Randomized Search Cross-Validation (CV) approach is used to fine-tune the model parameters and optimize performance. After conducting a thorough comparison with the most advanced techniques available, we have determined that our suggested approach surpasses the performance of the current methods