Research on the User Behaviour Game Analysis of Social Network

Rumour Propagation Based on the Weibo

Hailan Lan

Chengdu No.7 High School International Department, Chengdu, Sichuan, 610095, China

Keywords: Social Media, Weibo Rumor Propagation, Game Theory, User Behavior, Multi-Group Game-Theoretic

Model.

Abstract: “Rumor propagation on social media poses significant challenges to public trust and social stability.” In

today’s digital world, rumour propagation on social media has posed significant challenges to public trust and

social stability. Weibo, China’s dominant social media platform, accelerates rumor propagation through its

openness and real-time features, leading to social panic, trust crises, and economic losses. This paper

innovatively develops a multi-group game-theoretic model that classifies users into spreaders, refuters, and

ordinary users, analyzing their payoff functions and strategic choices through Nash Equilibrium theory to

uncover the potential influence factors of rumor spread. The study demonstrates that algorithmic

recommendation mechanisms, users’ psychological characteristics, and the response efficiency of debunking

systems constitute the three pivotal factors affecting rumor propagation. Accordingly, we propose an

integrated solution framework: restructuring platform incentives to modify user payoffs, implementing media

literacy programs, enhancing rumor-refuting systems for rapid response, and recalibrating algorithms to

reduce misinformation exposure. This research not only extends the application of game theory in social

media studies but also establishes a theoretically grounded and practically viable approach for platform

governance, providing actionable insights for improving social media governance and mitigating the negative

impacts of rumors.

1 INTRODUCTION

As the social network develops by leaps and bounds,

Weibo, as one of the largest social media platforms in

China, has become a significant channel for

information dissemination. However, the democracy

of knowledge and resources also makes rumor

propagation easier. Until 2025.3.14, the Chinese

Internet rumor-refuting platform Weibo has released

13097 posts about clarifying rumors, which even

ignored the unfound rumors. The spread of fake

information can not only cause social panic but can

also severely harm commercial brands, government

credibility, and individuals’ reputations. For instance,

The cover of Hong Kong-based Next magazine said,

“Bawang causes cancer” and claimed the product

contained dioxane. The whole article was merely

based on “expert speculation” alone, and the listed

company with annual sales of HK $1.7 billion lost

HK $2.4 billion. Therefore, investigating the users’

https://orcid.org/0009-0005-1148-0999

behavior in rumor spreading on the internet,

especially from a multi-group game-theoretic

perspective, is of significant practical importance and

urgency.

This paper selects Weibo as an example and a

research object to analyze participants’ acting

strategies(e.g., spreading, refuting, silence) when

they face rumors based on a game theory. It can not

only provide an iconic example and theoretical

foundations for rumor control on general platforms

but also offers valuable insights for national

authorities and enterprises in formulating information

management strategies, showcasing crucial social and

commercial values.

In the field of online rumor propagation, many

scholars have already explored the topic from various

perspectives, for example:

Li, Ma, and Fang investigated the impact of

emotion types and intensity on rumor spreading,

Lan, H.

Research on the User Behaviour Game Analysis of Social Network Rumour Propagation Based on the Weibo.

DOI: 10.5220/0013803600004708

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 2nd International Conference on Innovations in Applied Mathematics, Physics, and Astronomy (IAMPA 2025), pages 5-11

ISBN: 978-989-758-774-0

finding negative feelings have a more significant

influence on the evolution of rumors. But, they

mainly focused on sentimental aspects, lacking the

overall strategy analysis (Li, Ma, and Fang, 2023).

Yi, Liu, and Wang proposed the Twin-SIR model

to analyze the unsatisfactory speed of clarifying

rumors and introduce a “rumor dispeller” to improve

it. But, they did not deeply consider the game-

theoretic aspects of participants’ strategies (Yi, Liu,

and Wang, 2021).

Li and Liu set up the ET game model to find an

equilibrium but did not combine it into empirical data,

particularly social media (Li and Liu, 2016).

Liu and Liu developed a tripartite evolutionary

game model based on game theory. This consists of

the insider, the media, and the government regulatory

agencies (Liu and Liu, 2021).

Liu, Wang and Ouyang modeled all online users

are represented as a node in one of five compartments.

The transition dynamics are obtained as a system of

ordinary differential equations, and the system is

described using an edge-based formulation (Liu,

Wang and Ouyang, 2022).

Although much research has been conducted on

rumor propagation on the internet, most studies focus

on dynamic models, sentimental factors, or

improvement of present rumour detection systems;

few scholars have deeply investigated the interaction

of users’ behaviour strategies and application in

reality from a game-theoretic perspective. This paper

aims to fill this gap by constructing a multi-group

game-theoretic model to analyse user behaviour and

its impact on rumour propagation.

In order to achieve this, filling this gap, the following

structure of this paper is presented:

Firstly, the establishment of a multi-group game-

theoretic model is completed, categorizing

participants into spreaders, refuters, and ordinary

users and analyzing their strategy choices (e.g.,

spreading, refuting, silence) and payoff functions,

which is based on the real data from Weibo.

Secondly, the Nash Equilibrium will be solved to

explore different behavior strategies under

equilibrium and their influence on rumor

transmission.

Thirdly, parameter analysis and sensitivity testing

are conducted to examine the specific effect of

different factors (e.g., spreading benefits, refuting

cost) on rumor spreading.

Finally, the rumor-preventing strategies based on

the results will be proposed to offer the theoretical

foundations for rumor control on social platforms.

2 CASE DESCRIPTION

2.1 Introduction of Weibo as a Typical

Social Media Platform

Weibo, as the counterpart of Twitter, is one of the

most prevailing social media platforms in China, with

over 605 million monthly active users until 2023. It

enable people around the world to communicate

instantly and easily, democratizing variable

information and updating users on current affairs.

With the characteristics of openness and frequent

interaction, Weibo has become a hotspot for both

legal information dissemination and the transmission

of rumors. This platform’s real-time nature and large

user base make it particularly vulnerable to the rapid

spread of fake news, which can lead to critical social

and economic consequences.

2.2 Overview of Rumor Propagation on

Weibo

As a common phenomenon on Weibo, rumor

propagation is often exaggerated by high user

engagement and fast dissemination speed. According

to the “2023 Weibo ‘We-Media’ Governance Report”,

until 2023, Weibo has already handled over 27000

rumors and over 428000 fake accounts. These rumors

cover a wide range of aspects, including medical

health, food security, and social science, influencing

people quietly.

2.3 Main Characteristics of Rumor

Propagation on Weibo

There are several predominant features of rumor:

Fast speed and large scale: Since Weibo is a real-

time platform with the convenience and availability

of vast information, with just a single click, rumors

can spread there at a rate never seen before. Within

hours, a single rumor can reach millions of users,

making it challenging to stop once it starts to spread.

Conformity of user behavior: People frequently

participate in rumor spreading on Weibo merely

because they are curious, afraid, or want to contribute

eye-catching content. Therefore, rumors are further

spread via the platform’s algorithm, which gives

priority to popular subjects.

Impact on society and economy: Rumors on

Weibo have the potential to cause social unrest, harm

to people’s reputations, and financial losses, along

with other not-mentioned negative consequences. For

instance, the “Bawang causes cancer” event, which

IAMPA 2025 - The International Conference on Innovations in Applied Mathematics, Physics, and Astronomy

was mentioned before, has resulted in the loss of HK

$2.4 billion to the company.

According to the relevant research, the rumors

also tend to possess periodic spikes, different from the

single prominent sipes in terms of the temporal aspect,

illustrating the repeated characteristic of rumors; the

fake news is also inclined to form a “Bigfoot”

diffusion network rather than “Summize” diffusion

network, demonstrating the blindness and uni-

directivity of rumors (Kwon, Cha, Jung, Chen &

Wang, 2013). The spread of rumors on Weibo is a

complicated problem that presents serious difficulties

for the site and its users. Consequently, social media’s

dynamic nature necessitates ongoing efforts to

address this issue adequately.

3 ANALYSIS OF THE PROBLEM

3.1 Severe Influences of Rumor

Propagation

3.1.1 Social Panic and Trust Crisis

The rapid spread of rumors often leads to social unrest,

especially on prevailing social platforms like Weibo.

For example, during the COVID-19 pandemic in

2020, false information about the origin and treatment

of the virus spread widely on social media, resulting

in large-scale social panic and distrust in the

government and medical institutions (Bukhari, 2020).

Many people were already swayed by the rumors

when official sources tried to provide clarification,

which resulted in resistance and a skeptical attitude

towards official comments. Because it damages the

credibility of reliable sources and makes future

communication attempts more difficult, this crisis of

confidence may have long-term consequences. The

psychological impact of such rumors can also lead to

a breakdown in social cohesion, as individuals

become more suspicious of each other and less

willing to cooperate with authorities. These

declinations on government credibility not only

destabilize society but may also lead to public

resistance to official information, further

exacerbating the chaos in rumor dissemination.

3.1.2 Damage to Individual, Group, or

Brand Reputation

The blinded transmission of fake news can also cause

severe consequences to related people or brand

reputation. For instance, in 2022, there was a common

rumor that a “nurse was playing on her phone while

rescuing a child.” However, the nurse was actually

using her phone to call a doctor for help with the

rescue. The rumor-creators purposefully made up

false information to gather attention, which was

extremely deceptive, and took advantage of the

important roles that nurses and volunteers play in

preventing and controlling epidemics (Global Times,

2022). Despite the clarifications on the multi-

platforms, the impact of the rumor still exists and

even spreads. Such reputational damage not only

affects the short-term individuals’ mental health, the

credibility of medical workers, or the economic

interest of relevant corporations but may also have

long-term implications for their development. The

damage to reputation is often difficult to quantify but

can have lasting effects on both personal and

professional lives.

3.1.3 Fluctuation in Economy and Market

Rumor propagation can also trigger the whole

economy and market to become more unstable. Like

the “Bawang causes cancers” fake news, which is

mentioned before, a single false claim leads to a loss

of HK$2.4 billion for a listed company, causing

significant losses to investors. In addition to

disrupting businesses’ regular operations, this

economic volatility may also have an adverse impact

on the entire industry. In certain instances, rumors

may result in regulatory scrutiny, legal disputes, and

higher operating costs for businesses attempting to

repair the harm. The financial impact of rumors is not

restricted to specific businesses; it can also impact

entire sectors, causing wider economic instability.

This fake information about certain corporations can

lead to drastic fluctuations in stock prices and market

shares, undermining investor confidence and market

stability, resulting in a ripple effect on the entire

industry. In 2012, following political unrest in the

Maldives, a rumor spread claiming the country was

unsafe for tourists. Despite government clarifications,

the rumor led to massive cancellations of travel plans

and a sharp decline in hotel bookings. This incident

highlights the devastating impact of rumors on the

tourism industry, causing significant economic losses

for the Maldives in a short period.

3.1.4 Psychological Impact on Individuals

In addition to the negative social and economic

effects, rumors can have a significant psychological

effect on people. False information spreading can

make people anxious, stressed, and afraid,

particularly when it comes to health or personal safety.

For instance, during the COVID-19 pandemic,

Research on the User Behaviour Game Analysis of Social Network Rumour Propagation Based on the Weibo

widespread panic and irrational behavior, including

hoarding medical supplies and avoiding healthcare

institutions, were caused by rumors about the virus’s

lethality and the efficacy of specific therapies

(Bukhari, 2020). Vulnerable groups, such as the

elderly or people with underlying mental health issues,

may be most affected psychologically. Rumors have

an emotional cost that can result in long-term

psychological stress that impairs a person’s general

well-being and quality of life. According to related

research, rumor-spreading behavior is also more

significantly influenced by negative emotions (e.g.,

anger and disappointment), as the recipient is more

likely to spread rumors to express emotions or seek

empathy. This type of emotion-driven

communication behavior is especially prevalent in

social networks, which exacerbates the further

transmission of fake news (Kim & Bock, 2011).

3.2 Problems of Rumor Propagation on

Weibo

3.2.1 Uncontrollability of Information

Dissemination

As an open social platform, Weibo’s speed and scale

of diverse information dissemination are difficult to

control. A single rumor can reach millions of users

within hours, and the platform’s algorithm further

accelerates its spread. Although Weibo has already

implemented measures such as content moderation

and rumor-debunking systems, its response speed and

coverage are still insufficient to defeat the rapid speed

of false information fully. The fact that a rumor may

have reached a large amount of users by the time it is

detected and addressed makes it difficult to control.

This uncontrollability is a serious problem since it

permits rumors to proliferate unchecked, causing

extensive harm before they can be successfully

challenged.

3.2.2 Complexity of User Behavior

Participants’ behaviors in rumor propagation are

varied, ranging from passive sharing driven by

curiosity or fear to active sharing driven by personal

interest. Some users may share rumors without

verifying their authenticity simply because the

content aligns with their beliefs or interests, while

others may actively broadcast incorrect information

to get attention, gain followers, or even profit

monetarily. According to related research, the variety

of user behavior depends on the user’s information

perception(e.g., Number of followers and social

network activity) and activity, individuals’

psychological aspect(e.g., Profit-seeking psychology

or Herd mentality), and the competitive mechanism

of rumor and anti-rumor information (Xiao, Chen,

Wei et al., 2019). Another research study also pointed

out that people who hesitate to join in the propagation

of rumors can even exacerbate their speed (Hu, Pan,

Hou and He, 2018). This complexity makes it

challenging to predict the motivations behind rumor

propagation, which, therefore, further increases the

difficulty of rumor control.

3.2.3 Limitations of Rumor-Refuting System

Despite the diverse measures taken by Weibo to deter

the dissemination of rumors, the limitation of its

rumor-refuting system still exists. According to

related research, the speed and scale of refuting

information are often less than the rumor itself, and

the former is more likely to be confined within the

scientific chamber without being exposed to a large

number of ordinary people, which results in limited

efficiency. Additionally, the results of the research

showed that some users have low trust in rumor-

refuting information and even resist it. This resistance

can be due to a variety of factors, including pre-

existing biases, distrust of official sources, or the

influence of echo chambers within social networks

(Zollo, Bessi, Del Vicario et al., 2017). In some cases,

debunking efforts may even have the opposite impact,

as users who are deeply invested in a particular rumor

may view the debunking as an attempt to suppress the

truth, leading to further resistance and skepticism.

3.2.4 Algorithmic Amplification of Rumors

The platform’s algorithm’s function in spreading

misleading information is one of the biggest obstacles

to regulating rumor propagation on Weibo. The

algorithm gives priority to content that receives a lot

of likes, shares, and comments. The algorithm is more

likely to promote rumors than true facts since rumors

frequently provoke powerful emotional responses. As

a result of this algorithmic bias, rumors spread more

quickly since they become more visible when more

people share them. Recording to a relevant research,

videos with a high level of involvement (likes,

comments, shares) are typically recommended by the

social media’s algorithm. The rumor may spread

more quickly if the algorithm pushes it to more users

and creates a lot of interaction, especially if the

information is controversial. Second, the algorithm

might distribute the item to more users faster if it is

produced when users are active, which would

accelerate the spread of the rumor. Additionally, users

IAMPA 2025 - The International Conference on Innovations in Applied Mathematics, Physics, and Astronomy

still tended to employ algorithmic hashtags (such as

#fyp, #foryou) to boost visibility, even though the

study indicated that utilizing popular hashtags did not

directly enhance the number of video plays. The

algorithm may mistake the rumored content for

popular content and recommend it to additional

viewers if it utilizes these tags. Lastly, because people

don’t fully comprehend the algorithm, they try

different tactics to cater to it. Because of this

uncertainty, rumor content may be chosen by

algorithms and disseminated using a variety of tactics

(e.g., high interaction, contentious headlines, etc.)

(Klug, Qin, Evans and Kaufman, 2021). To solve this

problem, social media algorithms’ content

prioritization must be fundamentally rethought,

placing more weight on accuracy and dependability

than just engagement numbers.

4 SUGGESTIONS

4.1 Optimizing Platform Governance

Through Incentive Mechanisms

The rapid and large-scale spread of rumors on Weibo

stems from the platform’s open nature and

algorithmic amplification. To mitigate this, Weibo

should adopt a multi-layered governance approach by

using the multi-group game-theoretic model.

4.1.1 Construction of the Multi-Group

Game-Theoretic model

Participants: Spreaders, Refuters, and Ordinary users

Strategies: Spreaders: Spreading or Not Spreading;

Refuters: Refuting or Not Refuting; (P.S. Ordinary

users choose silence as the default strategy and do not

include it in the calculation of Nash Equilibrium.)

Parameter Setting:

Spreader’s payoff:

Successful spread: +B (social capital, e.g.,

increase the fan base).

Being refuted: −C (penalty, e.g., account

suspension).

Refuter’s payoff:

Successful refute: +R (platform reward).

Cost of refuting: −E (time effort).

Construction of the matrix, as shown in Table:

Table 1: Payoff Matrix of Rumor Propagation Participants

Refuters:

Refuting

Refuters: Not

Refuting

Spreaders:

readin

(B-C, R-E) (B, 0)

Spreaders:

Not Spreading

(0, -E) (0, 0)

4.1.2 Calculation of Nash Equilibrium

From Spreaders’ perspective, firstly, suppose

Refuters choose “Refuting”, Spreaders will get (B –

C) for “Spreading”, and 0 for “Not Spreading”.

Therefore, their optimal strategy is to choose

“Spreading” if (B−C)> 0; otherwise, the optimal

strategy is to choose “Not Spreading”.

Secondly, suppose Refuters choose “Not

Refuting”, Spreaders will get B for “Spreading”, and

0 for “Not Spreading”. Therefore, their optimal

strategy is always choosing “Spreading” if B > 0.

From Refuters’ perspective, firstly, suppose

Spreaders choose “Spreading”, Refuter will gets (R –

E) for “Refuting”, and 0 for “Not Refuting”.

Therefore their optimal strategy is to choose

“Refuting” if (R – E) > 0; otherwise, the optimal

strategy is to choose “Not Refuting”.

Secondly, suppose Spreaders choose “Not

Spreading”, Refuters’ payoff is 0, regardless.

Therefore, their optimal strategy is always choosing

“Not Refuting” to save cost.

Consequently, we can conclude 2 Nash

Equilibrium from the analysis above:

Firstly, when B > 0 and (R-E) <= 0, Spreaders

choose “Spreading”, and Refuters choose “Not

Refuting”.

Secondly, when (B-C) <= 0 and (R-E) > 0,

Spreaders choose “Not Spreading”, and Refuters

choose “Refuting”.

4.1.3 Application in Real Cases

The game-theoretic model suggests that modifying

user payoffs can effectively reduce rumor

propagation. Therefore, Weibo should modify these

payoffs in its incentive structure to lessen the spread

of rumors:

Firstly, reducing the benefits of spreading rumors

(B). Applying algorithmic demotion to remove

unconfirmed content from recommendation feeds and

hot topics reduces its visibility. Additionally,

implementing account penalties such as temporarily

suspending or shadow-ban accounts that repeatedly

share false information decreases their influence.

Research on the User Behaviour Game Analysis of Social Network Rumour Propagation Based on the Weibo

Secondly, increasing the costs of spreading

rumors (C). Using Fact-Checking Alerts to

automatically flag suspected rumors with warnings

(e.g., “This post conflicts with verified reports”).

Thirdly, rewards for refuting(R) should be

boosted. Offering tangible benefits (e.g., badges,

monetization opportunities) to users who actively

debunk rumors. In addition, allowing credible

debunking accounts to earn advertisements’ revenue

or tipping. Also, amplifying official corrections to

prioritize posts from verified sources (e.g., health

agencies) in users’ feeds.

Lastly, the effort to debunk (E) should be lowered.

Establishing One-Click Fact-Checking, such as

integrating tools like reverse image search and AI-

generated summaries of debunking articles.

By applying these measures, users are more

inclined to stop spreading and even debunking rumors;

the detection of fake news can also develop by leaps

and bounds.

4.2 Enhancing Public Awareness and

Critical Thinking

To alleviate the amplification of rumors from herd

mentality, several measures should be implemented

to increase media literacy:

First of all, constructing the interactive pop-ups.

Show quick alerts (such as “Have you checked this

with reliable sources?”) when people try to share

content that hasn’t been validated.

Secondly, gamify rumor detection by introducing

quizzes that test users’ ability to identify

misinformation. For example, users can complete

interactive quizzes on spotting fake news, with top

scorers earning badges and rewards such as platform

credits or exclusive features.

Thirdly, collaborating with Educators. Platforms

should cooperate with schools to promote digital

literacy campaigns targeting younger users.

Through the implementation of these measures,

rumors can be eliminated from their roots.

4.3 Improving the Rumor-Refuting

System

The inefficiency of the current system is caused by

poor credibility and delays. Among the solutions are:

Firstly, implement fact-checking in real time. For

example, AI can be used to identify possible rumors,

closing the “detection gap quickly”.

Secondly, use transparent procedures: Make the

process of verifying assertions publicly available (e.g.,

“This was debunked using CDC data”).

Thirdly, it involves participation at the grassroots

level. Related authorities should train volunteers to

become “rumor detectors” to spot local false

information promptly.

Consequently, by using the above solutions, the

present rumor-refuting system of Weibo could be

improved and provide more efficient and useful

services to users.

4.4 Algorithmic Adjustments to Curb

Rumor Transmission

Since Weibo’s engagement-driven algorithm

unintentionally spreads rumors, the reformation

should involve the following measures:

Firstly, the platform should prioritize the accuracy

signals, like ranking content, by using “trust scores”

(e.g., past accuracy of the poster).

Secondly, add brief refutation tips before users

confirm that they want to share websites. (e.g., “Read

this debunking article before sharing?”).

Thirdly, construct user customization. Let users

filter out unverified topics or mute frequent rumor

spreaders.

Through these algorithmic adjustments, the

platform can significantly reduce the unintentional

spread of rumors.”

5 CONCLUSION

This study provides a comprehensive analysis of

rumor propagation on Weibo, identifying key factors

and proposing actionable solutions from a game-

theoretic perspective. The analysis identifies several

critical issues, including complex user motivations

such as herd mentality and profit-seeking. These

motivations stem from users’ pursuit of economic

benefits and social recognition. Additionally,

algorithmic amplification prioritizing viral but

unverified content and inefficiencies in the rumor-

refuting system (e.g., delayed responses, low

credibility) are also involved in these issues. By

constructing a multi-group game model, the research

categorizes participants into spreaders, refuters, and

ordinary users, analyzing their strategy choices based

on payoff functions. Nash Equilibrium solutions

reveal two scenarios: rumor outbreaks when refuting

costs exceed benefits and rumor suppression when

incentives align. Proposed suggestions include

redesigning platform incentives, enhancing public

media literacy, optimizing fact-checking mechanisms,

and algorithmic reforms. These suggestions are aimed

at reducing the spread of rumors by adjusting

IAMPA 2025 - The International Conference on Innovations in Applied Mathematics, Physics, and Astronomy

participants’ payoff and improving the efficiency of

refuting rumors.

This study offers practical and social value. For social

media platforms like Weibo, our findings suggest

actionable measures to mitigate rumor risks, such as

penalizing unverified content and rewarding

debunkers. These measures aim to adjust user payoffs,

improve debunking efficiency, and reduce rumor

propagation. Policymakers may leverage the model to

design regulations (e.g., penalty adjustments) to

disrupt the rumors that lead to harmful consequences.

Additionally, by advocating digital literacy programs,

the research contributes to regaining public trust and

social cohesion, addressing challenges caused by the

transmission of false information during significant

events like COVID-19.

A limitation lies in its reliance on secondary data (e.g.,

published reports), lacking primary user behavior

validation. Future studies could: (1) collaborate with

platforms to analyze user logs for precise parameter

calibration (e.g., quantifying “B” and “ R”); (2)

compare cross-cultural differences in rumor-

spreading strategies; (3) integrate sentiment analysis

to explore emotional drivers of equilibrium shifts.

Addressing these limitations will enhance the

practical applicability of academic findings.

REFERENCES

Bukhari, W. 2020. Role of social media in COVID-19

pandemic. The International Journal of Frontier

Sciences, 4(2), 59–60.

https://doi.org/10.37978/tijfs.v4i2.144

Global Times. 2022, July 18. Nurse checking her phone

while rescuing patient arouses controversy. Retrieved

March 19, 2025, from

https://www.globaltimes.cn/page/202207/1270783.sht

Hu, Y., Pan, Q., Hou, W., & He, M. 2018. Rumor spreading

model with the different attitudes towards rumors.

Physica A: Statistical Mechanics and its Applications,

502, 331–344.

https://doi.org/10.1016/j.physa.2018.02.096

Kim, J., & Bock, G. 2011. A study on the factors affecting

the behavior of spreading online rumors: focusing on

the rumor recipient’s emotions.

Klug, D., Qin, Y., Evans, M., & Kaufman, G. 2021. Trick

and please: A mixed-method study on user assumptions

about the TikTok algorithm. In Proceedings of the 13th

ACM Web Science Conference 2021 (WebSci’ 21) (pp.

84–92). Association for Computing Machinery.

https://doi.org/10.1145/3447535.3462512

Kwon, S., Cha, M., Jung, K., Chen, W., & Wang, Y. (2013,

December). Prominent features of rumor propagation in

online social media. In 2013 IEEE 13th International

Conference on Data Mining (ICDM) (pp. 1103–1108).

IEEE. https://doi.org/10.1109/ICDM.2013.61

Li, M., & Liu, F. 2016, January. Game theory-based

network rumor spreading model. In 2016 International

Conference on Network and Information Systems for

Computers (ICNISC) (pp. 89–94). IEEE.

https://doi.org/10.1109/ICNISC.2016.029

Li, Y., Ma, J., Fang, F., & Jiang, Y. 2023. How emotion

type and intensity affect rumor spreading. International

Journal of Modern Physics C, 34(6).

https://doi.org/10.1142/S0129183123500833

Liu, C., & Liu, F. 2021, October. Tripartite evolutionary

game analysis of online rumor supervision based on

stakeholder theory. In 2021 IEEE 5th Information

Technology, Networking, Electronic and Automation

Control Conference (ITNEC) (pp. 580–583). IEEE.

https://doi.org/10.1109/ITNEC52019.2021.9586906

Liu, W., Wang, J., & Ouyang, Y. 2022. Rumor transmission

in online social networks under Nash equilibrium of a

psychological decision game. Networks and Spatial

Economics, 22, 831–854.

https://doi.org/10.1007/s11067-022-09574-9

Xiao, Y., Chen, D., Wei, S., et al. 2019. Rumor propagation

dynamic model based on evolutionary game and anti-

rumor. Nonlinear Dynamics, 95, 523–539.

https://doi.org/10.1007/s11071-018-4579-1

Yi, J., Liu, P., Wang, Z., & L, W. 2021. Research on twin-

SIR rumor spreading model in online social network.

Journal of Intelligent & Fuzzy Systems: Applications in

Engineering and Technology, 40(4), 5863–5874.

https://doi.org/10.3233/JIFS-189426

Zollo, F., Bessi, A., Del Vicario, M., Scala, A., Caldarelli,

G., Shekhtman, L., Havlin, S., & Quattrociocchi, W.

2017, July 24. Debunking in a world of tribes. PLOS

ONE. https://doi.org/10.1371/journal.pone.0181821

Research on the User Behaviour Game Analysis of Social Network Rumour Propagation Based on the Weibo