본문 바로가기
  • Home

Analyzing Immersive Media Art Tourism Experiences Using BERTopic

  • Journal of Internet of Things and Convergence
  • Abbr : JKIOTS
  • 2026, 12(1), pp.111~120
  • Publisher : The Korea Internet of Things Society
  • Research Area : Engineering > Computer Science > Internet Information Processing
  • Received : January 20, 2026
  • Accepted : February 20, 2026
  • Published : February 28, 2026

Seong-bin Ko 1 KANG JUHYUN 1

1제주대학교

Accredited

ABSTRACT

This study analyzes visitor experiences and emotional responses to the immersive media art exhibition “Nohyung Supermarket” in Jeju by examining 1,166 Google Maps reviews. We collected 2,000 reviews in reverse chronological order and retained 1,166 valid entries after removing missing texts and noise. Using BERTopic-based topic modeling and KcBERT-based sentiment analysis, we identified experiential themes in user-generated content and examined their relationships with sentiment and ratings. Texts were preprocessed and embedded using a RoBERTa-based Korean multitask pretrained model, with UMAP and HDBSCAN applied within BERTopic; sentiment probabilities were computed with KcBERT-base and cross-analyzed by topic. The results highlight sensory immersion, photo spots, family-oriented experiences, and indoor alternative tourism, with strong positive sentiment for immersive and emotionally engaging spatial experiences. In contrast, admission fees, limited space, and circulation inconvenience showed higher negative sentiment, indicating that emotional satisfaction and practical considerations jointly shape evaluations. These findings suggest that media art exhibition management should address both experiential appeal and physical/economic conditions

Citation status

* References for papers published after 2024 are currently being built.