网络信息传播动力学建模研究

对网络传播机理建模是利用或控制网络传播的基础。网络信息传播系统具有复杂的动力学特征,其中的个体行为具有适应性。采用社会物理学研究思路,借鉴物理学振动传播的概念,对网络传播系统中各动力学因子进行了类比定义。根据动力学因子对网民传播行为的影响机理,建立具有随机性和适应性的网民信息传播微观动力学模型,使微观个体行为与宏观的系统状态关联起来。在此基础上构建仿真模型,通过仿真实验分析了网络传播系统S型函数的非线性演化特征,为类似系统的建模提供了新思路。

Abstract：

Modeling the mechanism of information diffusion is the basis of network diffusion utilizing and controlling. The diffusion of information on the Internet is with dynamical complexity, and the individual behavior is adaptive. Adopting the thinking in social physics, the dynamic factors in the information diffusion system on the analog of the spread of vibration in physics was defined. Then the microscopic dynamic model of net user’s randomicity and adaptability behavior was quantitatively built based on the mechanism of dynamic factor taking effect on the behavior of net users. In this way, the microscopic behavior was associated with system evolution at macro level. Simulation software was developed based on those models. The S_shape nonlinear evolution characteristics of information diffusion on the Internet with computing experiments were analyzed. This solution may be used in modeling similar systems.

Dynamical Modeling of Information Diffusion on Internet

Modeling the mechanism of information diffusion is the basis of network diffusion utilizing and controlling. The diffusion of information on the Internet is with dynamical complexity, and the individual behavior is adaptive. Adopting the thinking in social physics, the dynamic factors in the information diffusion system on the analog of the spread of vibration in physics was defined. Then the microscopic dynamic model of net user’s randomicity and adaptability behavior was quantitatively built based on the mechanism of dynamic factor taking effect on the behavior of net users. In this way, the microscopic behavior was associated with system evolution at macro level. Simulation software was developed based on those models. The S_shape nonlinear evolution characteristics of information diffusion on the Internet with computing experiments were analyzed. This solution may be used in modeling similar systems.