K-means聚类算法在SIR传染病模型中的应用研究

Research on the Application of K-means Clustering Algorithm in SIR Infectious Disease Model

基于SIR传染病模型,建立了具有K-means聚类算法的SIR元胞自动机模拟模型.通过对分别服从高斯分布和随机均匀分布的两类初始感染源的分析与模拟,给出了疾病感染半径与隔离半径对疾病传播的影响.结果显示:在两种不同类型的初试分布下,感染者的最大值分别与疾病感染传播半径和隔离半径呈正相关与负相关关系,感染者数量随时间的变化率亦呈现相同的变化规律.初始数据的不同分布类型只影响这种正负相关关系的增速.研究结果可为控制和消除传染病提供有效合理的隔离措施,为卫生部门提供防控传染病的理论支持.

Based on the SIR infectious disease model,the SIR cellular automata simulation model with k-means clustering algorithm is established in this paper.Through the analysis and simulation of two types of initial infectious sources that obey gaussian distribution and random uniform distribution respectively,the influence of infection radius and isolation radius on disease transmission is given.Under the first distribution of two different types,a maximum of infected with disease transmission radius and isolation of radius were positively correlated with negative correlation,the number of infected people over time and the rate of change of rendering the same change rule.The different distribution of the initial data type affects only the positive and negative correlation between growth.The results could be used to control and eliminate infectious disease provide effective reasonable isolation measures,for the health sector to provide theoretical support for prevention and control of infectious diseases.