Temporal and Content Coupling Analysis of Social Media User Behavior

Researchers from the field of social and information networks have published a paper on arXiv (2604.27530) analyzing how temporal dynamics and content selection jointly influence news consumption. Using two large real-world datasets—MIND and Adressa—the team developed a multi-scale framework that reveals hierarchical temporal patterns. At the macroscale, Fourier modeling identifies clear circadian rhythms in user activity. At the mesoscale, session intervals follow a power-law distribution with exponent α≈1, indicating bursty behavior. At the microscale, within-session action counts and inter-action intervals follow exponential distributions with λ≈0.3 and λ≈0.02 respectively.

The content analysis shows that clicks are mainly driven by historical interests, but this dependence weakens as content diversity increases. Temporal-content coupling further indicates that users' historical interests dominate during active time periods. The study also identifies preference groups: timeliness and entertainment-oriented users click more frequently and rely heavily on historical interests, whereas diversified users click less and are more sensitive to content diversity. These findings have implications for recommendation systems and content platforms seeking to optimize user engagement by accounting for both timing and content preferences.