基于复杂网络的观点动力学的研究

发布时间：2018-06-19 23:49

本文选题：统计物理 + 观点动力学　；参考：《北京邮电大学》2017年硕士论文

【摘要】：近年来,采用自然科学的研究方法研究复杂的社会群体现象和行为引起了人们的广泛兴趣。针对社会群体中观点的传播行为,观点动力学把观点传播抽象为物理模型,通过计算机编程在复杂网络上进行数值模拟的方式,进行观点演化的研究。观点动力学主要研究观点的演化、形成、以及达成共识的过程。物理模型主要分为离散观点动力学模型和连续观点动力学模型。本文主要对改进后Deffuant连续观点动力学模型(引入排斥机制与周期观点)在一个环形网络上进行了观点的演化研究。排斥机制使两个人的观点彼此远离。周期观点是指把观点值具有周期性。我们发现排斥机制的引入会使邻居间的观点产生极化现象;周期观点的引入会使系统演化出一种全新的“全局一致”,在这种一致状态中所有节点的观点值并不一致,但相邻节点间的观点差值非常小,而且当系统大小趋于无穷时任意两个邻居节点之间的观点差是趋近于零的。Deffuant模型同时引入周期观点和排斥机制后,观点的演化结果既依赖置信区间又依赖演化参数。排斥机制和周期观点的共同作用可能会为我们进一步研究时间相关的观点动力学提供一种思路。研究结果对人们深入理解观点的形成和演化的内在机理有一定的意义;针对大规模人群更好的决策和讨论过程,引导和控制社会舆论的传播和社会共识的达成同样具有重要的理论价值。
[Abstract]:In recent years, the use of natural science research methods to study complex social groups phenomenon and behavior has attracted widespread interest. In view of the communication behavior of ideas in social groups, viewpoint dynamics abstracts the dissemination of ideas as a physical model, and studies the evolution of views by means of numerical simulation on complex networks by computer programming. Viewpoint dynamics mainly studies the evolution, formation, and the process of reaching consensus. The physical model is mainly divided into discrete viewpoint dynamics model and continuous viewpoint dynamics model. In this paper, the modified Deffuant continuous viewpoint dynamics model (including the exclusion mechanism and the periodic viewpoint) is studied on a ring network. The exclusion mechanism keeps the views of two people away from each other. A periodic point of view refers to the periodicity of the value of the point of view. We find that the introduction of the exclusion mechanism leads to polarization of the viewpoint between the neighbors, and the introduction of the periodic point of view makes the system evolve into a new "global consistency" in which the views of all nodes are not consistent. However, the difference of views between adjacent nodes is very small, and when the system size tends to infinity, the difference of views between any two neighbor nodes is near zero, and the periodic viewpoint and the exclusion mechanism are introduced into the Deffuant model. The result of evolution depends on both confidence interval and evolution parameters. The interaction of the exclusion mechanism and the periodic viewpoint may provide a way for us to further study the time-related viewpoint dynamics. The results of the study have certain significance for people to deeply understand the internal mechanism of the formation and evolution of views, and for the better decision-making and discussion process of the large population, Guiding and controlling the spread of public opinion and reaching social consensus also have important theoretical value.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：O157.5

【参考文献】