摘要
针对经典的传染病动力学模型在处理COVID-19时存在的问题,提出一种基于改进的时变SEIR模型的流行病传播模型.首先,在引入潜伏者仓室的基础上,优化模型参数;然后,充分利用多项式回归与多层感知机算法的非线性逼近能力和动态学习优化能力,对模型的时变参数进行拟合;最后,提出基于最优参数的疫情传播建模方案,并将其应用于COVID-19疫情传播建模中.多组实验表明,提出的方法较好地拟合了疫情传播过程,取得了最小的均方根误差,适合对疫情传播进行建模.
Aiming to the problems that classical epidemic dynamics models in dealing with COVID-19,this paper proposed an epidemic transmission model based on improved time-varying SEIR model.First,it optimized the parameters of the model on the basis of the introduction of the exposed compartment.Then,it fitted the time-varying parameters of the model by making full use of the nonlinear approximation ability and dynamic learning optimization ability of polynomial regression and multi-layer perceptron algorithm.Finally,it put forward the epidemic transmission modeling scheme based on the optimal parameters,and applied it to the epidemic transmission modeling of COVID-19.Groups of experiments show that the proposed method can well fit the epidemic transmission process and obtain the minimum root mean square error,which is suitable for epidemic transmission modeling.
作者
李静
高媛
黄家妹
LI Jing;GAO Yuan;HUANG Jia-mei(Computer Department, Xinzhou Teachers University, Xinzhou Shanxi 034000, China;School of Data Science and Technology, North University of China, Taiyuan 030051, China;Department of Electrical Engineering and Computer Science, The University of Tennessee, Knoxville State of Tennessee 37996, America)
出处
《中北大学学报（自然科学版）》
CAS
2021年第6期486-494,共9页
Journal of North University of China(Natural Science Edition)
基金
山西省本科教学质量提升工程项目(J2020291)。
关键词
COVID-19
多项式回归算法
多层感知机算法
疫情传播建模
Corona Virus Disease 2019(COVID-19)
polynomial regression algorithm
multi-layer perceptron algorithm(MLP)
epidemic transmission modeling
