Enhancing Botnet Detection in IoT-Enabled Healthcare Networks Using a Hybrid Deep Learning Framework

The increasing prevalence of technologies has enabled innovative infrastructures powered by IoT, with smart cities, smart Agro-technology in the agriculture field, smart grids and smart healthcare systems. Security is the main challenge in every field. Among these, botnet attacks pretense to be a critical hazard to IoT-based networks. This article analyses various botnet attacks on the IoT-based healthcare networks and proposes a hybrid model of machine learning and deep learning algorithms to predict the attacks. The proposed framework integrates a Bi-directional Long Term Memory for the extraction of the features with the Random Forest classifier for the final classification. The model was evaluated using the N-BaIoT datasets, achieving model accuracy of 99.99% with the binary classification while attaining an 84.7% F1 score and 97.56% accuracy for the N-BaIoT datasets. The proposed model demonstrates the effectiveness in the detection of various botnet attacks.