95598 Network Traffic Prediction Based on CNN-BiLSTM-Attention Optimization Algorithm

Abstract

In order to accurately predict short-term fluctuations in network traffic and improve the traffic monitoring performance of the 95598 customer service hotline network, this article combines convolutional neural networks (CNN), bidirectional long short-term memory networks (BiLSTM), and attention mechanisms (Attention) to input various traffic related influencing factors, and constructs a deep network model for CNN BiLSTM Attention network traffic prediction. Multiple traffic related influencing factors are input, Comprehensive construction of a deep network model for small-scale network traffic prediction. The experimental results show that compared with traditional shallow neural network prediction models and deep network LSTM prediction models, the method proposed in the article not only achieves higher accuracy in short-term traffic prediction, but also achieves better results in longer time series traffic prediction.