Sina Weibo's "Super Topic Communities" Data

This included posts and fan activity records between January 27, 2023, and February 2, 2023. This period was chosen as it coincides with the Chinese New Year holiday, a time of heightened fan activity.

During this period, data on each fan’s basic information and original posts were also collected. To maintain a balance between high-level and low-level fans, and considering the potential for outliers in the number of posts, followers, and followings, the Isolation Forest algorithm (Liu et al., 2008) in Python was applied to remove anomalous outliers from both groups. This resulted in 9,248 remaining users’ data (high-level fans: 4,608; low-level fans 4,640).