.png){kind=icon}
.png){kind=icon}
Return
A Guided Topic Dissemination Model Based on Cognitive Differences
.png){kind=icon}
PRE
AI
DOI：10.1109/TCSS.2025.3602669.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In social networks, studying the propagation patterns of guided topics helps to understand their dissemination trends, thereby adjusting the direction of topic evolution. To address the issue of users’ cognitive closure in topic propagation, this article proposes a guided topic propagation model based on cognitive differences. First, considering the complex factors influencing topic propagation, this article explores the benefits of a multifactor model through multiple linear regression analysis. It proposes a method to measure guided topics influence, accounting for both internal and external factors, to effectively quantify their impact on propagation. Second, considering the symbiosis and antagonism of different guided topics, this article examines the impact of game theory on user behavior, proposing a driving mechanism based on cognitive game theory. It quantifies the forces behind message forwarding, analyzes the relationships between topics, and explores the cognitive game states of users. Finally, considering the heterogeneity of user cognition and the advantages of cognitive difference theory in modeling state transitions, along with the strengths of mean-field theory for large-scale information dissemination, this article incorporates known, rational, and irrational states into the susceptible-infectious-recovered (SIR) model. An susceptible-exposed-rational information/irrational-immune guided topic propagation model is then constructed to analyze topic propagation patterns. Experiments demonstrate that this model effectively analyzes the impact of cognitive differences on the dissemination of guided topics, exploring the dissemination patterns of guided topics within social networks.
Keywords:
Cognitive game theory
guided topics
social networks
user cognition
Journal
IF:
4.9
Papers:
577
Citations:
6.8K