A social diffusion model with an application on election simulation.

Area of Science:

• Social Network Analysis
• Computational Social Science
• Political Science

Background:

• Traditional opinion diffusion models struggle with election scenarios due to pre-existing individual attitudes and the multi-faceted nature of political opinions.
• Existing models often simplify opinions to single values, failing to capture complex preferences for multiple candidates in electoral contexts.

Purpose of the Study:

• To design and develop a novel opinion diffusion model capable of handling the complexities of electoral opinion formation.
• To provide an empirical approach for modeling the interflow and formation of crowd opinions in elections.

Main Methods:

• Development of a new diffusion model that accommodates intrinsic individual attitudes and multi-candidate preferences.
• Simulation of opinion diffusion on a network constructed from a publicly available bibliography dataset.
• Comparative analysis against established models like Independent Cascade.

Main Results:

• The proposed model consistently demonstrated superior performance compared to existing diffusion models.
• The model successfully simulated electoral opinion dynamics, offering insights into crowd opinion formation.

Conclusions:

• The developed model offers a more intuitive and accurate approach to understanding opinion diffusion in electoral settings.
• This work provides a valuable tool for investigating electoral issues and crowd behavior through simulation.