Research On HIV Transmission Model Based On Agent Social Circle Network

This paper established a model which combined multi-agent systems and social-circle networks; we have used this model to study the simulation process of HIV spread in high-risk groups, such as intravenous drug users (IDU). It could provide references for HIV prevention and control decision-making.The prevalence of the HIV has been one of the most challenging problems in public health all over the world, and extensive researches on it have been carried out to seek effective measures to inhibit HIV epidemic. The prevention and control of HIV in China have made some progress, but further control is still needed in key region of high prevalence and key populations at high risk. Therefore, in-depth study of HIV transmission’s internal mechanism is significant for the government to take effective measures to prevent and control the spread of the HIV.This paper has done some various simulation researches of virus simulation based on multi-agent social-circles networks.First, An agent-based complex network simulation model was proposed by studying the mechanism of combining agent system and complex network. A social-circle network was used to describe the underlying social networks that would be required in many social simulations, and agent incorporated dynamics to simulate the population migrations and interpersonal relationships changing.Second, An agent-based dynamical social circles network simulation model was proposed by studying the dynamic propagation of HIV among intravenous drug users. This model was used to account for syringes-shared HIV transmission, the social influence, floating population and the interaction of these factors, it embodied the responsiveness and adaptability of agent.Third, this paper established a heterosexual network model by using an evolutionary method, in this model, agents had their own preferences in sexual partner choosing, and the relationship between sexual partner network and social network was built. It provided basis for hybrid model in which there were more than one networks.This paper used models that combined the agent system and the complex network. It solved the traditional HIV models’problem of normalization and immobilization. Agent was used to act as the HIV spreading individual in the society, the individual autonomy behaviors were built into the model, and the interpersonal relationship dynamic was simulated. By analyzing the simulation results of the model, the effects of various factors in the process of the propagation were evaluated. Based on these conclusions, governments’HIV control strategies can be more targeted and effective and the spread of the HIV can be better control and the inhibition.