Abstract:
The increasing prevalence of dynamic malware in IoT devices necessitates the development of robust detection techniques. This research proposes a novel approach utilizing a deep boosted CNN-RNN architecture, specifically LSTM, for accurate and efficient dynamic malware detection in IoT environments. By combining the strengths of convolutional neural networks (CNN) and recurrent neural networks (RNN), the model captures both spatial and temporal features, enabling effective analysis of malware behavior. To enhance the model's performance, a Bayesian hyperparameter optimization technique is employed to find the optimal weights, resulting in improved accuracy and reduced false positives. The proposed approach achieves outstanding results, with an accuracy of 0.9915, precision of 0.985, recall of 0.992, and an F1 score of 0.9885. These metrics demonstrate the model's ability to accurately classify malware instances while maintaining a low false positive rate. The significance of this research lies in its optimized approach, which effectively addresses the challenges of dynamic malware detection in IoT. The integration of Bayesian hyperparameter optimization enables the model to adapt and achieve superior performance, thereby enhancing the security and reliability of IoT devices. This research contributes to the advancement of malware detection techniques, providing valuable insights for the development of robust security measures in IoT environments.
Page(s):
327-327
DOI:
DOI not available
Published:
Journal: Abstract Book on International Conference on Food and Applied Sciences (ICFAS-23) 3-5 August 23, Volume: 0, Issue: 0, Year: 2023
Keywords:
IoT security
,
Malware Detection
,
Malware detection mechanismin IoT
,
Novel Neural Network
,
Deep Boosted CNN
,
IoT malware detection