基于带有非理性因素的协调博弈的舆情传播研究

发布时间：2018-04-24 07:09

本文选题：从众心理 + 非理性因素　；参考：《吉林大学》2015年硕士论文

【摘要】：随着科技的不断发展，借助网络，通讯等途径传播舆情信息对社会的危害正日益扩大。舆情信息是指那些在社交网络中快速传播的信息，这些信息可以是谣言，不法言论等等，也可以是一些带有正能量的信息。由于科技的进步使得这些舆情信息可以快速传播并造成较大的社会影响。由于网络具有虚拟性，使得追查舆情信息的源头变得很困难，这加大了控制舆情的难度。通常控制舆情传播的方法是限制其在网络中传播，但是阻止信息在网络中传播这本身违背了网络本身的优点，同时会引起人们的反感，因此本文提出了另一种控制舆情传播的方法。本文提出了通过控制网络中个体对舆情信息的认知来达到控制舆情传播的目的。在本文的模型中，舆情信息可以随意在网络中传播，而研究的重点放在了个体的对于舆情的认知过程上。只要网络中的个体对于舆情信息有一个正确的认识，那么就会减轻舆情信息所带来的社会影响。本文采用了博弈论中的协调博弈模型来模拟网络中个体的认知过程。协调博弈是要满足预期行为一致的博弈。协调博弈本身描述的就是博弈参与者协商的过程，这和网络中个体讨论舆情信息的情况相似。本文还加入了带有非理性因素的协调博弈模型对邻居之间交互进行建模，重点研究了在从众心理的非理性因素的作用下，网络中个体对于舆情信息认知的演化过程。这里的非理性并非通常所说的感性，而是博弈参与者选择自身策略时不考虑收益，根据其他条件来选择策略。博弈中的非理性因素是近些年所提出的，它拟补了传统博弈论本身的局限，即博弈论假设参与者为纯理性人，只考虑在博弈过程中所能获得的收益，进而选取策略。非理性因素有很多，本文中的非理性因素为从众心理，它是指博弈参与者不考虑收益，而是跟随其邻居选择策略的情况，选择多数邻居所选择的策略。同时，本文还研究了一些其他因素，如冒险系数、初始偏好对个体认知过程的影响。冒险系数是对选择风险策略的一种激励，很多情况下由于一些传统观念等原因，正确的认知并不能被广泛认同，这时就需要激励那些能够选择正确认知的人，以达到整体认知趋于正确的目的。可以说冒险激励是一种控制舆情传播的方法。在现实世界中，由于每个人所处的生活环境不同，社会经历不同，对于同一实物看法也会不同，初始偏好就是用来模拟这种情况的。随着不断的博弈讨论，初始偏好的影响也会逐渐降低。实验结果表明非理性因素不能改变总体的策略选择偏好，但能对部分个体的策略选择产生影响，，使得网络中出现选择不同策略的群体，且两种群体不能互相侵入。冒险系数，初始偏好可以控制总体策略选择，对舆情传播的控制具有一定的作用。针对舆情传播在复杂网络上的研究有助于了解网络中舆情信息传播的过程以及相关个体对于舆情信息的交互和认知规律，能够进一步揭示真实世界社交网络的舆情传播规律，通过这些规律能过为社交网络上的政策制定提供一定的理论基础以及为遏制舆情带来的危害提供相应的策略。
[Abstract]:With the continuous development of science and technology, the public opinion information is spreading to the society by means of network and communication. Public opinion information is the information which is spread quickly in the social network. These information can be rumor, illegal speech and so on. It can also be some positive energy information. It makes it difficult to trace the source of public opinion because of the virtual nature of the network, which makes it more difficult to control public opinion. Generally, the way to control public opinion transmission is to restrict its spread in the network, but it is against the network itself to prevent the spread of information in the network itself. The advantages will also cause people's resentment. Therefore, this paper proposes another way to control the spread of public opinion.
This paper puts forward the purpose of controlling public opinion transmission by controlling the individual's cognition of public opinion information in the network. In this model, the public opinion information can be spread at random in the network, and the focus of the research is on the individual cognition process of public opinion. As long as the individual in the network has a correct recognition of public opinion information. Knowledge will reduce the social impact of public opinion information.
This paper uses the coordinated game model in the game theory to simulate the cognitive process of the individual in the network. The coordinated game is a game to satisfy the consistent expected behavior. The coordination game itself describes the process of negotiation among the players, which is similar to the individual discussion of public opinion information in the network.
In this paper, a coordinated game model with irrational factors is added to model the interaction between neighbors, and the evolution process of the individual's cognition of public opinion information in the network is studied under the effect of the irrational factors of the crowd. The irrationality here is not usually said by the perceptual, but when the game participants choose their own strategies. The irrational factor in the game is put forward in recent years without considering the income. The irrational factor in the game is proposed in recent years. It has made up the limitations of the traditional game theory itself, that is, the game theory assumes that the participants are purely rational, only considering the gains that can be obtained in the game process, and then select the strategy. There are many irrational factors and the non rational factors in this paper. Sexual factor is a herd mentality. It refers to the game participants who do not consider revenue but follow their neighbors' choice strategy and choose the strategy chosen by most neighbors.
At the same time, some other factors, such as the risk factor, the influence of the initial preference on the individual cognitive process. The risk factor is an incentive to choose the risk strategy, and in many cases, due to some traditional ideas, the correct cognition is not widely recognized, and then it is necessary to encourage those who can choose the right cognition. It can be said that risk incentive is a way to control the spread of public opinion. In the real world, because the living environment of each person is different, the social experience is different, the opinion of the same kind will be different, the initial preference is used to simulate this situation. With the continuous game discussion, The effect of initial preference will also gradually decrease.
The experimental results show that irrational factors can not change the overall strategic choice preference, but it can affect the selection of strategies for some individuals, which makes the network appear to select groups of different strategies, and the two groups can not invade each other. The risk coefficient, initial preference can control the selection of the general strategy, and the control of public opinion transmission is certain. The role.
Research on the spread of public opinion on complex networks helps to understand the process of dissemination of public opinion information in the network and the interaction and cognitive laws of relevant individuals on public opinion information. It can further reveal the law of public opinion dissemination in real world social networks. Through these laws, it can provide a certain reason for policy making on social networks. On the basis and provide corresponding strategies to curb the harm brought by public opinion.

【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：O157.5;O225

【参考文献】