基于神经网络的流感大数据分析

从互联网大数据中挖掘传染性疾病的早期征候、传播趋势、演变规律，对于疫情监测及相关政策制定具有重要的现实意义。人工神经网络在区域传染性疾病监测和趋势分析中的应用主要侧重于传染病早期征候大数据对于疫情局势影响的回归分析，但对于传染病早期征候及疫情演变历史的关系的研究依然缺失。通过构建流感征候指标体系，利用循环神经网络对流感历史数据进行时序建模，并结合注意力机制挖掘不同特征对流感的关联影响，实现区域流感的准确监测和趋势分析。

Neural network-based analysis of influenza big data

It is of important realistic significance to monitor epidemic diseases and work out related policies by mining the big data of their early syndrome, spreading trend, and evolution rules on Internet. Artificial neural network has been applied in monitoring regional infectious diseases and their trend analysis, especially in regression analysis of the effect of early syndrome big data on the situation of epidemic infectious diseases. However, studies on their evolution history are quite few. It can accurately monitor regional influenza and analyze its trend to establish neural network-based time sequence model of influenza historical data by developing the system for influenza syndrome indicators in combination with attention mechanism in mining the effect of different characteristics on the association of influenza.