面向舆情治理的信息管理机制演化博弈模型

doi:10.11959/j.issn.2096-109x.2023086

摘要/Abstract

摘要：

在线社交网络平台中存在的冲突信息加剧了群体观念对立，需要科学合理的管理机制来实现舆情治理。提出面向舆情治理的信息管理机制演化博弈模型，通过建立动态奖惩策略，自动化地实现冲突抑制，激活网络自我净化能力。通过引入演化博弈理论，构建基于累积前景理论的社交网络用户信息冲突演化博弈模型，使用数值模拟的方法研究 9 个约束条件下的 18 种理论应用情景的演化结果，分析了双方行为互动演化规律，得出约束条件五中博弈双方理性对待信息冲突的结论。通过建立博弈分析的系统动力学模型，对动力学演变进行分析，并在满足约束条件五的前提下对模型参数进行了详细的灵敏度分析，并为动态信息管理机制提供了实用化的策略建议。所提模型首次对利益驱动、从众心理等多种因素在舆情演化中的相互作用进行建模，演化分析结果与谣言优势传播规律相符合，证实了所提模型的合理性。所提模型为社交网络中恶意信息控制系统的开发和设计提供了重要借鉴，为舆情治理的研究实践提供了新的思路。

关键词: 演化博弈, 信息管理, 信誉, 社交网络, 舆情治理

Abstract:

Conflicting information in online social networks, often stemming from opposing views, can lead to rumors and the fragmentation of society.A reputation-based dynamic information management mechanism was proposed, adopting a dynamic reward and punishment strategy to decrease unnecessary information conflict so as to activate self-purification of network.By utilizing evolutionary game theory, a cumulative-prospect-based evolutionary game model of social network information conflict was constructed to validate the efficacy of the proposed mechanism.Numerical simulations were conducted to analyze the evolutionary trends of 18 different theoretical application scenarios under 9 constraints.The results demonstrate that, under constraint 5, both sides of the game tend to exhibit relatively rational behavior towards conflicts.Furthermore, dynamic evolution analysis and sensitivity analysis of the model parameters were performed while satisfying constraint 5, providing practical strategic suggestions for the dynamic information management mechanism.This model represents a novel attempt to capture the interaction of multiple factors, such as interest-driven behavior and conformity psychology, in the evolution of public opinion.Moreover, the model successfully simulates well-known patterns of rumor dissemination, confirming its validity.The proposed model offers valuable insights for the development and design of malicious information control systems in social networks and provides new ideas for the research and implementation of public opinion governance.

Key words: evolutionary game, information management, reputation, social network, public opinion governance

中图分类号:

TP393

刘淑婷, 马颖华, 陈秀真. 面向舆情治理的信息管理机制演化博弈模型[J]. 网络与信息安全学报, 2023, 9(6): 102-115.

Shuting LIU, Yinghua MA, Xiuzhen CHEN. Evolutionary game model of information management mechanism for public opinion governance[J]. Chinese Journal of Network and Information Security, 2023, 9(6): 102-115.

图/表 12

图1

表1

图2

表2

表3

表4

不同均衡点对应雅克比矩阵的行列式和迹Table 4 Rank and trace of jacobi matrix corresponding to different equilibrium points"

均衡解	$\det (J)$	$tr (J)$
F₁	$\begin{array}{l} λ^{2} E_{1} (K_{1} + C_{1})^{β} \\ (K_{2} + C_{2})^{β} \end{array}$	$\begin{array}{l} λ [E_{1} (K_{1} + C_{1})^{β} + \\ (K_{2} + C_{2})^{β}] \end{array}$
F₂	$λ^{2} E_{1} D_{1} H_{2}$	$- λ (E_{1} D_{1} + H_{2})$
F₃	$λ^{2} E_{1} D_{2} H_{1}$	$- λ (E_{1} H_{1} + D_{2})$
F₄	$λ^{2} E_{1} I_{1} I_{2}$	$λ (E_{1} I_{1} + I_{2})$
F₅	$\begin{array}{l} x^{} y^{} (1 - x^{}) (1 - y^{}) \cdot \\ (J_{11}^{'} \|_{x = x^{}} J_{22}^{'} \|_{y = y^{}} - \\ J_{12} \|_{y = y^{}} J_{21} \|_{x = x^{}}) \end{array}$	$\begin{array}{l} x^{} (1 - x^{}) J_{22}^{'} \|_{y = y^{}} + \\ y^{} (1 - y^{}) J_{11}^{'} \|_{x = x^{}} \end{array}$

表4

表5

图3

图4

图5

图6

图7

参考文献 29

[1]	ALLPORT G W , POSTMAN L . The psychology of rumor[R]. 1947.
[2]	靖鲲鹏, 岑怡 . 社交网络舆情下的二元信息竞争 SH2IR 模型及仿真研究[J]. 情报杂志, 2022,41(1): 117-124.
	JING K P , CENG Y . SH2IR model and simulation of binary information competition under social network public opinion[J]. Journal of Intelligence. 2022,41(1): 117-124.
[3]	黄炜, 余辉, 李岳峰 ,等. 网络舆情事件演化的仿真实践研究[J]. 现代情报, 2017,37(8): 65-73.
	HUANG W , YU H , LI Y F ,et al. Simulation practice of the network public opinion evolution[J]. Journal of Modern Information, 2017,37(8): 65-73.
[4]	PAL S , HILL A , RABEHAJA T ,et al. A blockchain-based trust management framework with verifiable interactions[R]. 2021.
[5]	CHEN Y Z , LI Q L , WANG H . Towards trusted social networks with blockchain technology[J]. Social and Information Networks, 2018,18(1): 2796.
[6]	CHEN J , CHEN C , SONG Q ,et al. Spread mechanism and control strategies of rumor propagation model considering rumor refutation and information feedback in emergency management[J]. Symmetry, 2021,13(9): 1694.
[7]	WANG Z , CHEN A . On ISRC rumor spreading model for scale-free networks with self-purification mechanism[J]. Complexity, 2021(8): 1-9.
[8]	MO L , GUO S . Consensus of linear multi-agent systems with persistent disturbances via distributed output feedback[J]. Journal of Systems Science and Complexity, 2019,32(3): 835-845.
[9]	ZHAO L , QIU X , WANG X ,et al. Rumor spreading model considering forgetting and remembering mechanisms in inhomogeneous networks[J]. Physica A:Statistical Mechanics and its Applications, 2013,392(4): 987-994.
[10]	ASKARIZADEH M , LADANI B T . Soft rumor control in social networks:Modeling and analysis[J]. Engineering Applications of Artificial Intelligence, 2021,100:104198.
[11]	SIMON S , APT K R . Choosing products in social networks[C]// 8th Workshop Internet Netw Econ(WINE). 2012: 100-113.
[12]	MACHINA M J . Expected utility hypothesis[C]// Utility and Probability. 1990: 79-95.
[13]	WANG B , GE C , FU L ,et al. DRIMUX:dynamic rumor influence minimization with user experience in social networks[J]. IEEE Transactions on Knowledge and Data Engineering, 2017(10): 1.
[14]	TSCHIATSCHEK S , SINGLA A , GOMEZ RODRIGUEZ M ,et al. Fake news detection in social networks via crowd signals[C]// Companion Proceedings of the Web Conference 2018. 2018: 517-524.
[15]	NGUYEN N P , YAN G , THAI M T ,et al. Containment of misinformation spread in online social networks[C]// Proceedings of the 4th Annual ACM Web Science Conference. 2012: 213-222.
[16]	WEN S , JIANG J , XIANG Y ,et al. To shut them up or to clarify:restraining the spread of rumors in online social networks[J]. IEEE Transactions on Parallel ＆ Distributed Systems, 2014,25(12): 3306-3316.
[17]	SRINIVASAN S , DHINESH B . A social immunity based approach to suppress rumors in online social networks[J]. International Journal of Machine Learning and Cybernetics, 2021,12(2): 1-16.
[18]	WU Y , HUANG H , ZHAO J ,et al. Using mobile nodes to control rumors in big data based on a new rumor propagation model in vehicular social networks[J]. IEEE Access, 2018:1.
[19]	TVERSKY A , KAHNEMAN D . Advances in prospect theory:cumulative representation of uncertainty[J]. Journal of Risk and Uncertainty, 1992,5(4): 297-323.
[20]	SKYRMS B . Dynamics of conformist bias[J]. The Monist, 2005,88(2): 260-269.
[21]	LAHBIB A , TOUMI K , LAOUITI A ,et al. Blockchain based trust management mechanism for IoT[C]// 2019 IEEE Wireless Communications and Networking Conference (WCNC). 2019.
[22]	LI L , KURTHS J , YANG Y ,et al. Prevention and trust evaluation scheme based on interpersonal relationships for large-scale peer-to-peer networks[J]. Mathematical Problems in Engineering, 2014,2014(3).
[23]	YANG Z , YANG K , LEI L ,et al. Blockchain-based decentralized trust management in vehicular networks[J]. IEEE Internet of Things Journal, 2018:1.
[24]	CHOI D , CHUN S , OH H ,et al. Rumor propagation is amplified by echo chambers in social media[J]. Scientific Reports, 2020,10(1): 1-10.
[25]	GINTIS H . Game theory evolving:a problem-centered introduction to modeling strategic behavior[M]. Princeton: Princeton University Press, 2000.
[26]	FRIEDMAN D . On economic applications of evolutionary game theory[J]. Journal of Evolutionary Economics, 1998(8): 15-43.
[27]	许伦辉, 骈宇庄, 林永杰 ,等. 基于累积前景理论的无信控路段人车通行冲突博弈分析[J]. 中国公路学报, 2022,35(1): 18.
	XU L H , PIAN Y Z , LIN Y J ,et al. Game analysis of conflicts between pedestrians and vehicular traffic on unsignalized road sections based on cumulative prospect theory[J]. China Journal of Highway and Transport. 2022,35(1): 18.
[28]	吴俊明, 刘炎 . 德法兼治:网络谣言之二元治理——兼论证券市场“网络谣言”的规制[J]. 证券法苑, 2016,18(2): 374-393.
	WU J M , LIU Y . Governing both germany and law:dual governance of internet rumors——also on the regulation of "internet rumors" in the securities market[J]. Securities Law Review, 2016,18(2): 374-393.
[29]	李国武 . 谣言实现的社会机制及对信息的治理[J]. 社会, 2005(4): 143-155.
	LI G W . Social Mechanism of rumor realization and information governance[J]. Society, 2005(4): 143-155.

变量		含义	说明
	x	支持者选择转发信息的概率	0≤x≤1
支持者	T₁	支持者的信誉值	T ₁＞0
	K₁	支持者传播信息的基础收益	K ₁＞0
	C₁	支持者争议基金	C ₁＞0
	y	反对者选择举报信息的概率	0≤y≤1
反对者	T₂	反对者的信誉值	T ₂＞0
	K₂	反对者反馈信息的基础收益	K ₂＞0
	C₂	反对者争议基金	C ₂＞0

变量	含义	说明
α	收益状态下风险敏感度	0＜α＜1
β	亏损状态下风险敏感度	0＜β 1＜
λ	损失规避程度	λ＞ 1
γ	收益状态下决策函数的弯曲程度	0≤γ≤1
δ	亏损状态下决策函数的弯曲程度	0≤δ≤1

变量	含义	说明
μ₁	从众效应权重	0≤μ ₁≤1
P_{B 1}	信息转发热度	0≤P_B1≤1

约束条件编号	约束条件内容	从众	情景	沉默点	举报点	转发点	冲突点	中心点
一	D ₁＞0，I₁＞0	E ₁＞0	1		√	√
	D ₂＞0，I₂＞0	E ₁＜0	2					√
二	D ₁＞0，I₁＞0	E ₁＞0	3		√
	D ₂＜0，I₂＞0	E ₁＜0	4					√
三	D ₁＞0，I₁＞0	E ₁＞0	5		√
	D ₂＜0，I₂＜0	E ₁＜0	6				√
四	D ₁＜0，I₁＞0	E ₁＞0	7			√
	D ₂＞0，I₂＞0	E ₁＜0	8		√
五	D ₁＜0，I₁＞0	E ₁＞0	9					√
	D ₂＜0，I₂＞0	E ₁＜0	10		√
六	D ₁＜0，I₁＞0	E ₁＞0	11					√
	D ₂＜0，I₂＜0	E ₁＜0	12		√		√
七	D ₁＜0，I₁＜0	E ₁＞0	13			√
	D ₂＞0，I₂＞0	E ₁＜0	14		√
八	D ₁＜0，I₁＜0	E ₁＞0	15					√
	D ₂＜0，I₂＞0	E ₁＜0	16		√
九	D ₁＜0，I₁＜0	E ₁＞0	17				√
	D ₂＜0，I₂＜0	E ₁＜0	18		√