Social network influence based on SHIR and SLPR propagation models.

Area of Science:

• Computational Social Science
• Network Science
• Information Science

Background:

• The internet and social media accelerate information and emotion dissemination.
• Unintended negative outcomes arise from rapid emotional transmission online.
• Analyzing social network influence is crucial for understanding online behavior.

Purpose of the Study:

• To propose novel propagation models (Susceptible Hesitant Infected Removed and Susceptible Latent Propagative Removal) for social network influence analysis.
• To develop an emotional communication model incorporating news and public opinion effects.
• To investigate the impact of network structures on user behavior and emotion spread.

Main Methods:

• Utilized Susceptible Hesitant Infected Removed (SHIR) and Susceptible Latent Propagative Removal (SLPR) propagation models.
• Developed an emotional communication model to assess news and public opinion impacts.
• Conducted experiments analyzing node influence, network structures, and emotion spread dynamics.
• Validated the model using Baidu Index big data analysis.

Main Results:

• A significant relationship exists between emotion spreader/hesitant density and node influence across network configurations.
• Reduced node influence led to lower peaks in hesitators and disseminators.
• The model demonstrated high accuracy and reliability with minimal error (max 0.04) compared to Baidu Index data.
• Node influence directly correlates with the speed and extent of emotional propagation.

Conclusions:

• Node influence is a key factor in modulating emotional propagation on social media.
• The proposed models accurately predict and analyze emotion spread dynamics.
• Findings offer insights for managing online emotional environments and mitigating negative outcomes.