Identifying Factors Influencing Variety Show Popularity Through Social Media Data Analytics and Topic Modeling
Variety shows are TV programs that integrate diverse performance styles to entertain audiences. It has played a significant role in the global entertainment landscape for over 70 years. With the fast-growing social network, variety shows have evolved considerably and remain a popular option for leisure. However, the factors driving the widespread attraction of variety shows have yet to be fully explored. To address this gap, this study analyzed viewer comments on a mystery detective show, Who's the Murderer, using comments collected from Chinese TikTok. A total of 10,264 comments were analyzed using the topic modelling technique Latent Dirichlet Allocation (LDA). The model identified four main themes, i.e., Horror and Suspense, Humor and Rewatch Value, Emotional Resonance, and Overall Positive Feedback with a weighted percentage of each word in each session. The results reveal two key insights. On one hand, emotional engagement is instrumental in driving the show's popularity. Touching moments, humor, and even horror elements could foster a strong sense of resonance among the audiences. On the other hand, by examining the weight of the topics across each session, the study identified session-specific features that correspond to audience preferences. This result indicated the successful features of this show include the involvement of popular and famous stars; the quality of its story; and the humor and touching scenes. This work not only enhances our understanding of the factors that lead to the popularity of variety shows but also provides actionable insights for content creators and producers seeking to optimize content and enhance audience engagement.
Authors
- Long Xiong (ORCID: https://orcid.org/0000-0002-8702-1976)
Publication Details
- Published
- 2025-03-15
- DOI
- https://doi.org/10.1109/isec64801.2025.11147250
- Primary Topic
- Computational and Text Analysis Methods
- Type
- article
- Field-Weighted Citation Impact
- 0.00