基于包容性检验和神经网络的网络流量预测模型

网络流量具有高度复杂的非线性特征，采用单一预测模型往往难以达到理想的预测效果，为此，提出一种包容性检验和BP神经网络相融合的网络流量预测模型（ET-BPNN）。首先采用多个单一模型对网络流量进行预测，然后通过包容性检验，根据t统计量检验选择最合适的基本模型，最后采用BP神经网络对基本模型预测结果进行组合得到最终预测结果。实验结果表明，相对于单一模型以及传统组合模型，ET-BPNN更加准确刻画了网络流量变化趋势，各项评价指标均达到更优，为实现网络流量准确预测提供了更为科学的方法。

Network traffic prediction based on encompassing tests and neural network

Network traffic has nonlinear characteristics, a single prediction model is often difficult to achieve the ideal prediction effect, and therefore, a novel network traffic prediction model is proposed in this paper based on encompassing tests and BP neural network (ET-BPNN). Firstly, the network traffic is prediction by some single models, and then the most suitable single models are selected by encompassing tests according to t statistic test, finally BP neural network is used to combine the prediction results of single models to obtain the final prediction result of network traffic. The experimental results show that, compared with the single model and traditional combination models, the proposed model can more accurately describe the change trend of network traffic and obtain better indexes, it provides a more scientific method for network traffic prediction.