Аннотация
The rapid proliferation of Internet of Things (IoT) devices has ushered in unprecedented levels of connectivity, offering enhanced automation and data sharing across diverse domains. However, this surge in interconnectivity has brought about a significant challenge regarding cyber security threats and intrusions. To overcome this problem, we propose a Hybrid Bayesian Tree Fusion (HBTF) for predicting the detection system. These systems leverage the capabilities of Artificial Intelligence to predict and proactively mitigate potential security breaches. We gather the NSL-KDD data set for evaluating the proposed method for forecasting intrusion detection. The collected data are preprocessed using Z score normalization to eliminate the noise. Extract the feature from the preprocessed data using kernel-based independent component analysis (KICA). The experimental findings of the proposed method are evaluated by recall, precision, f1-score, and accuracy from the result. The HBTF is greater than the existing method. The integration of AI augments the capability of IoT networks to counter emerging threats, ultimately fostering a safer and more secure interconnected world.