基于不完全信息演化博弈模型的网络舆情传播羊群行为  

Herd behavior in the dissemination of public opinion on the internet based on evolutionary game model with incomplete information

作　　者：刘锦德[1] 刘咏梅[1] 

机构地区：[1]中南大学商学院,湖南长沙410012

出　　处：《国防科技大学学报》2013年第5期96-101,共6页Journal of National University of Defense Technology

基　　金：国家自然科学基金委创新研究群体项目(70921001);教育部博士点基金资助项目(20110162110065);湖南省哲学社会科学基金资助项目(12YBB271)

摘　　要：以网络舆情传播过程中大部分人对事实不了解为前提,在这部分群体内构建基于不完全信息环境的两人对称演化博弈模型。提出了两人对称博弈矩阵来决定个体策略的选择,建立了复制动态方程说明不同策略群体比例的演化方向,演化时引入个体的记忆长度,根据设定的交互规则更新各自的观点值及记忆列表。仿真分析模型表明,在此环境中很容易产生羊群行为,群体内的个体都选择盲从并且观点保持一致。此外,羊群行为的集聚产生受多种因素影响,主要与采取分析策略的交互利益、成本系数以及记忆长度有关。本文对科学分析引导网络舆情的传播有一定实际意义。The spread of public opinion across the Internet in China was proposed by constructing a two-player symmetric evolutionary game model with incomplete information, since most of the players have limited knowledge about the spread and exact nature of public opinion. The two- player symmetric game model was presented to decide the strategic choice of players, and the replicated dynamic equation was built to explain the evolution directions of different strategic groups. The memory length of players, updates views and memory list were introduced according to the interaction rules. The simulation result shows that it is easy for players to follow trends under such circumstances, leading to herd behavior and a world in which most players adopt the same view. Furthermore, herd behavior is affected by a series of factors ： most notably analysis-strategy based interactive interests, cost coefficients, and memory length. This research offers insights for practical application on how to analyze and guide the spread of public opinion on the internet.

关 键 词：两人对称演化博弈 复制动态方程 网络舆情 羊群行为 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]