A Semantic Structure Analysis of Aesthetic Perceptions in Short-Form Hair Styling Content: A Text Mining Approach

With the rapid development of digital media environments, short-form content platforms such as TikTok and Instagram Reels have become important channels for the diffusion of beauty trends and aesthetic information. In particular, short-form hair styling contents visually present style transformations, color changes, and styling techniques within a short time, influencing users’ aesthetic perceptions and style preferences. Despite this trend, previous hair beauty studies have focused mainly on survey-based approaches, while research using actual digital content data remains limited. Therefore, this study aimed to analyze the structural characteristics, meaning structure, and sentiment responses reflected in short-form hair styling contents distributed on TikTok and Instagram Reels. To achieve this purpose, text data extracted from captions, hashtags, and comments were analyzed using text mining techniques, including keyword frequency analysis, TF-IDF analysis, LDA topic modeling, semantic network analysis, platform comparison, and sentiment analysis. The results showed that the major keywords were centered on hair, style, color, and transformation, indicating that short-form hair contents are mainly organized around visual style change and color trends. Topic modeling identified four major themes: hair style transformation, hair color trends, styling tutorials, and professional hair services. These findings suggest that short-form hair contents function as an important medium for shaping contemporary aesthetic perception and beauty trend consumption.