RBoost-Net: A Hybrid Deep Learning Model for Efficient Cybersecurity Threat Detection to Preserve Privacy in IoT Devices

The exponential growth of Internet of Things (IoT) devices has led to an increasing concern for cybersecurity due to the vulnerabilities present in these interconnected systems. This paper presents RBoost-Net, a hybrid deep learning model that integrates ResNet for feature extraction and LightGBM for classification to efficiently detect cybersecurity threats in IoT environments. The RBoost-Net model was evaluated against nine existing machine learning and deep learning models, including CNN, SVM, Random Forest, Logistic Regression, XGBoost, Naive Bayes, Gradient Boosting Machine, ANN, and LightGBM. The evaluation metrics considered were accuracy, precision, recall, F1-score, AUC-ROC, training time, inference time, and false positive rate. RBoost-Net achieved superior performance across all metrics, with an accuracy of 95.32%, significantly outperforming the compared models. The model's ability to balance accuracy, speed, and false positive rate makes it an ideal candidate for real-time IoT cybersecurity applications. This hybrid approach demonstrates the potential of combining deep learning with boosting techniques for efficient and accurate threat detection. The paper also discusses the advantages of RBoost-Net in identifying various IoT threats, including Denial of Service (DoS), Distributed Denial of Service (DDoS), data breaches, and malware, and highlights its scalability and adaptability in diverse IoT environments.