Text Mining Analysis of Research Trends in K-Beauty

This study explores academic research trends in the field of Korean beauty, commonly known as K-beauty, by employing advanced text-mining techniques. As global interest in K-beauty products and culture has rapidly increased, academic discourse surrounding K-beauty has likewise expanded. To systematically analyze this trend, 369 Korean academic papers containing the term “K-beauty” were collected from the Research Information Sharing Service database. These texts underwent a rigorous preprocessing procedure that included tokenization, part-of-speech tagging, and stopword removal, ensuring the quality and relevance of the extracted data. Keyword analysis was conducted using both CountVectorizer and Term Frequency–Inverse Document Frequency techniques. These complementary methods enabled the identification not only of frequently occurring keywords but also of contextually significant terms that might otherwise be overlooked. The extracted keywords were visualized using word clouds to highlight their relative prominence. To further uncover underlying patterns and relationships among terms, a keyword co-occurrence network was constructed and analyzed using four centrality measures—degree, closeness, betweenness, and eigenvector centrality—calculated via Python’s NetworkX library and visualized with UCINET software. The results revealed that keywords such as “cosmetics,” “brand,” and “industry” held dominant positions across all centrality measures, underscoring their foundational role in K-beauty research. Notably, “development” and “strategy” stood out in betweenness centrality, indicating their importance as conceptual bridges connecting different thematic clusters. Eigenvector centrality further highlighted clusters of highly influential terms, suggesting the presence of core themes shaping academic discourse. The findings point to a robust and interconnected research network characterized by both structural cohesion and thematic diversity. Unlike prior studies that focused on specific cases or limited timeframes, this study provides a holistic, data-driven overview of the evolving academic landscape in K-beauty. The results offer valuable insights for scholars, practitioners, and policymakers interested in strategic research planning, topic identification, and cross-disciplinary collaboration within the field of beauty studies.