Computer Information Security Management System Based on Artificial Intelligence Technology

Computer Information Security Management System Based on Artificial Intelligence Technology

연구 분야: Networking

학회: 2024 Second International Conference on Networks, Multimedia and Information Technology (NMITCON)

Network attack highlights the harsh requirements for computer information security management system. This study focuses on the performance of Convolutional Neural Network (CNN), Recurrent Neural Network (RNN) and Gated Recursive Unit (GRU) in threat detection, throughput and resource consumption. In this experimental research design, we set up data preprocessing, model architecture design and evaluation indicators. We use standardized data sets to train and evaluate CNN, RNN and GRU models, focusing on threat detection rate, throughput and resource consumption, while achieving performance comparison. In the experiment, CNN achieved the highest threat detection rate, up to 99%, the lowest resource consumption, and the peak utilization rate of central processing unit (CPU) was 31%. In contrast, compared with CNN, RNN and GRU have lower threat detection rate, higher resource consumption and lower throughput. The peak throughput of CNN is 278Kbps, which is faster in data processing and network traffic.