Discrete time information diffusion in online social networks: micro and macro perspectives

Area of Science:

• Social Network Analysis
• Information Diffusion Modeling
• Computational Social Science

Background:

• Online social networks facilitate rapid information dissemination.
• Accurately modeling this spread is crucial for understanding public opinion.
• Existing models face challenges in capturing complex diffusion dynamics.

Purpose of the Study:

• To develop novel models for information diffusion in online social networks.
• To analyze diffusion from both micro and macro perspectives.
• To incorporate user availability and external interferences into diffusion models.

Main Methods:

• Developed a discrete-time bi-probability independent cascade model considering user online/offline states.
• Established a macro-level diffusion model integrating event interferences and temporal cumulative effects.
• Investigated micro-level factors influencing message diffusion and macro-level interferences.

Main Results:

• The proposed discrete-time bi-probability independent cascade model effectively simulates information spread.
• The macro-level diffusion model accurately captures cumulative effects and interferences.
• Experimental validation using real-world data supports the efficacy of the developed models.

Conclusions:

• The study provides robust models for understanding information diffusion in social networks.
• Incorporating user availability and external factors enhances diffusion modeling accuracy.
• The findings offer valuable insights for monitoring and managing online public opinion.