Competition for popularity and interventions on a Chinese microblogging site

Area of Science:

• Social Media Analytics
• Computational Social Science
• Information Science

Background:

• Microblogging platforms like Sina Weibo are crucial for information dissemination and public opinion formation.
• Trending topics are typically presented via ranking lists, reflecting user engagement and platform dynamics.

Purpose of the Study:

• To investigate public attention dynamics on Sina Weibo's Hot Search List (HSL).
• To characterize hashtag rank dynamics, including time spent, appearance times, rank diversity, and trajectories.
• To identify platform intervention patterns and their potential impact on public opinion.

Main Methods:

• Analysis of hashtag ranking dynamics on Sina Weibo's HSL.
• Characterization using time spent, appearance time, rank diversity, and ranking trajectories.
• Application of machine learning clustering to categorize rank trajectories.
• Development of a model to explain ranking anchoring effects.

Main Results:

• Circadian rhythms significantly affect hashtag popularity.
• Distinct categories of hashtag rank trajectories were identified using machine learning.
• Anomalies in ranking dynamics suggest platform provider intervention, specifically hashtag anchoring.
• International politics hashtags showed over-representation at anchored ranks on the HSL.

Conclusions:

• Hashtag popularity on microblogging platforms exhibits complex dynamics influenced by user behavior and platform algorithms.
• Platform provider interventions, such as anchoring, can manipulate the visibility of certain topics.
• The over-representation of international politics hashtags at anchored ranks raises concerns about potential public opinion manipulation.