Traffic Intrusion Detection of Medical Consumption Electronics in the Field of Medical Management Based on Integrated Learning

The medical and health industry is evolving from traditional single management to modern scientific and systematic management. From the perspective of medical management, medical consumer electronics have developed rapidly. Various medical consumer electronics have been widely used, and the security of medical electronic networks has gradually become the focus of researchers. Traffic intrusion detection is a network security system running at all times, and its importance is self-evident. Traditional intrusion detection mainly relies on expert experience, which leads to the limitations of intrusion detection technology. To address this issue, this work designs a medical consumer electronic traffic intrusion detection network via deep learning and integrated learning. First, this work proposes a classification model (NDR-BiLSTM) based on the combination of the nonlinear dimensionality reduction method and BiLSTM. Secondly, this work proposes a processing model for unbalanced data problems (SMOTE-IAdaboost). This model can promote classification effect of minority data in intrusion detection. SMOTE-IAdaboost uses NDR-BiLSTM intrusion detection model as a weak classifier in ensemble learning. This eventually forms an intrusion detection model that improves the ability of unbalanced data processing. Finally, this work conducts systemic experiments, the data verify the superiority of SMOTE-IAdaboost for medical consumer electronic traffic intrusion detection.