Data-driven Transmission Dynamics In Social Network And Its Applications

The spread of virus and information(behavior,public opinion,etc.)on the social network is affected by individual behavior,natural and social environment.At the same time,a large number of data will be generated along with the spread process.It is inevitable to use the actual data to study various kinds of spread on the social network.Therefore,based on the data of population statistics,disease monitoring and geographical location,combined with the transmission network structure and transmission mechanism,this paper studies the transmission dynamics on social network,and applies it to the specific disease and behavior transmission on social network,forming the basic framework of data-driven.Its main contributions are as follows:1)Taking H7N9 avian influenza and African swine fever as the research object,a new data-driven model that combine data,network structure and transmission mechanism is proposed.Based on the actual data,we constructed the transmission network of specific diseases,and established data-driven network transmission dynamics models to study the transmission law of diseases on social networks;2)Taking African swine fever as the research object,we used the data-driven model that only based on data to analyzing the spatial and temporal characteristic of African swine fever in mainland China by spatiotemporal statistical analysis method.3)We apply the data-driven model that combine data,network structure and transmission mechanism to the behavior transmission on the social network.Taking the marriage relationship as the research object,we established the dynamic model of approaching the marriage network,to study the evolution law of the marriage relationship on the marriage network.The main research contents and innovations are summarized as follows:(1)It solved the problem about infection source tracing of H7N9 avian influenza human cases.In order to determine the infection source of the fifth wave of H7N9 avian influenza human cases,we used JavaScript language to crawl the geographic location data of relevant places of live birds from Baidu map,and built the live bird transportation network;based on the discrete Markov process,we established the data-driven H7N9 avian influenza transmission dynamics model on the network;proposed the spatiotemporal backward detection and forward propagation algorithms on the directed weighted network.The most likely infection source of each human case of the fifth wave H7N9 avian influenza was detected,and it was found that in addition to the live poultry market,backyard poultry was an important infection source;the most likely route map was deduced,and it is found that H7N9avian influenza spread from the Yangtze River Delta and the Pearl River Delta to the West;the first arrival time of the virus was calculated,and the temperature characteristics of the first arrival time of the virus were analyzed,we get that that the risk of human infection with the in H7N9 virus was high under 9℃~19℃.The new data-driven modeling idea we proposed can provide new ideas for the study of the transmission of specific diseases on the real network.(2)The transmission risk of three main transmission routes of African swine fever in mainland China were determined.In order to assess the transmission risk of the three main transmission routes of African swine fever in mainland China,we used Python language to crawl the geographic location data from Baidu map and constructed the pig transportation network;based on the discrete Markov process,we established the data-driven transmission dynamics model of African swine fever on the network;we proposed the spatiotemporal backward detection and forward propagation algorithms on the semi-directional weighted network.The transmission risks of three transmission routes were analyzed,the results shows that the transmission risk of people and vehicles carrying virus were the largest,followed by the swills with virus,and the transmission risks of pig and pig product transportation were relatively small.The most possible transmission route map was deduced,and it was found that African swine fever spread from northeast to Southwest and then to West in mainland China.In addition,the infection risks in provinces were assessed at different times.(3)The spatiotemporal transmission characteristics of African swine fever in mainland China were revealed.In view of the rapid spread of African swine fever in mainland China from 2018 to 2019,based on the data of African swine fever cases and the total amount of pig farms in various provinces,we used spatiotemporal statistical analysis method is to analyze the spatiotemporal transmission characteristics of African swine fever in mainland China.We found the hot spots of African swine fever in mainland China,which are concentrated in some cities in Northeast and southwest China;Seven spatio-temporal clusters of African classical swine fever are determined,and the most likely spatio-temporal cluster was located in Buyi and Miao Autonomous Prefecture of QianNan in Guizhou Province and the cluster time was from June 19 to 25,2019.The first secondary cluster areas were covered five cities(Shenyang,Yingkou,Panjin,Anshan and Liaoyang)in Liaoning Province from August 1 to October 10,2018.The global and local transmission direction and speed of African swine fever in mainland China were given,it is found that the spatial transmission speed of the disease was slow from August to October 2018,and fast from February to March 2019.(4)Some effective suggestions were given to improve the marriage rate in China.For the current situation that the marriage rate is decreasing year by year in China,we put forward the new evolution mechanism of marriage with mutual introduction among friends,and established the pair approximation dynamics model of marriage network.Based on the marriage registration data in Beijing,the feasibility of the model is verified by the numerical solution of the model and the stochastic simulation on ER network.Through the sensitivity analysis,it is found that the introduction among friends can not only increase the threshold,but also increase the number of married people in the network.At the same time,it is found that increasing the number of same sex friends and balancing the sex ratio can also the married size and increase the marriage rate.