演化博弈视角下自媒体平台“信息茧房”生成机制研究

Research on the Generation Mechanism of"Information Co‐coon"of We-Media Platform from the Perspective of Evolution‐ary Game

随着人工智能、大数据、云计算等新信息技术的发展,信息传播进入新时代。算法凭借其优势,已成为自媒体平台个性化信息传播的核心工具。然而,这种算法驱动的信息传播方式也带来了一系列挑战,如信息茧房。在算法时代下,深入分析信息茧房的形成机制、探究信息茧房的困境的破解办法、规范信息传播是当务之急。本研究从演化博弈理论的角度探讨了自媒体平台推荐算法下信息茧房的形成机制。通过构建自媒体平台与用户之间的博弈模型,本研究分析了信息传播过程中平台和用户的策略选择变化如何导致信息茧房现象。研究发现,通过增加平台推荐异质信息的成本、提高用户接受异质信息的收益,以及政府对同质化推荐策略的惩罚,可以有效减少信息茧房的形成。本研究为规范自媒体推荐算法提供了策略建议,旨在促进信息的多样性和公平性,维护健康的网络信息生态。

With the development of new information technologies such as artificial intelligence,big data and cloud computing,information dissemination has entered a new era.With its advantages,algorithms have become the core tools of personalized information dissemination on we-media platforms.However,this algorithm-driven approach to information dissemination also brings its own set of challenges,such as the information cocoon.Therefore,in the era of algorithms,it is imperative to deeply analyze the formation mechanism of information cocoon,explore the solution to the dilemma of contemporary information cocoon,and standardize the information transmission.This study discussed the formation mechanism of information cocoon under the recommendation algorithm of we-media platform from the perspective of evolutionary game theory.By constructing an evolutionary game model between the we-media platform and users,this study analyzed how the change of platform and user's strategy choice in the process of information transmission leads to the phenomenon of information cocoon.It was found that the formation of information cocoon can be effectively reduced by increasing the cost of platform recommendation of heterogeneous information,improving the benefit of users receiving heterogeneous information,and punishing the homogenized recommendation strategy by the government.This study provided strategic suggestions for standardizing we-media recommendation algorithms,aiming at promoting the diversity and fairness of information and maintaining a healthy network information ecology.