Complex Network Traffic Prediction Method Under Graph Neural Network
摘要
Accurate and real-time network traffic prediction has an important role in networks, and it is also essential in traffic engineering and network control. The current prediction methods have problems with low prediction accuracy and inability to adapt to complex traffic. In order to achieve the purpose of efficient use of network resources, this article uses the GraphSAGE spatial model in graph neural networks (GNN) to predict complex network traffic. The results showed that the prediction accuracy rate of GNN is higher.