网络流量预测是网络安全领域重要的研究方向之一，精准预测网络流量的趋势和峰值，并针对现有信息安全系统发现网络中可能存在的安全问题做出提前预警。随着各传感器的大量部署，系统已拥有大量可用数据，但是缺乏行之有效的分析方法，为此本文通过深度学习的方式对网络流量预测建立模型，提出一种基于LSTM神经网络的流量预测模型，并与ARIMA模型比较验证LSTM网络模型具有更好的性能，验证了该模型在网络流量预测中的适用性与更高的准确性。 Network traffic prediction is one of the important research directions in the field of network security. It accurately predicts the trend and peak value of network traffic, and provides early warning for existing information security systems that may find security problems in the network. With the large-scale deployment of various sensors, the system has a large amount of available data, but lacks effective analysis methods. For this reason, this paper establishes a model for network traffic prediction through deep learning, and proposes a traffic prediction model based on LSTM neural network , and compared with the ARIMA model to verify that the LSTM network model has better performance, and verify the applicability and higher accuracy of the model in network traffic prediction.
展开▼
机译：网络流量预测是网络安全领域重要的研究方向之一，精准预测网络流量的趋势和峰值，并针对现有信息安全系统发现网络中可能存在的安全问题做出提前预警。随着各传感器的大量部署，系统已拥有大量可用数据，但是缺乏行之有效的分析方法，为此本文通过深度学习的方式对网络流量预测建立模型，提出一种基于LSTM神经网络的流量预测模型，并与ARIMA模型比较验证LSTM网络模型具有更好的性能，验证了该模型在网络流量预测中的适用性与更高的准确性。 Network traffic prediction is one of the important research directions in the field of network security. It accurately predicts the trend and peak value of network traffic, and provides early warning for existing information security systems that may find security problems in the network. With the large-scale deployment of various sensors, the system has a large amount of available data, but lacks effective analysis methods. For this reason, this paper establishes a model for network traffic prediction through deep learning, and proposes a traffic prediction model based on LSTM neural network , and compared with the ARIMA model to verify that the LSTM network model has better performance, and verify the applicability and higher accuracy of the model in network traffic prediction.
展开▼
