텍스트 마이닝 기법을 활용한 코로나-19 이전과 이후의 화장품 소비 문화 및 기호가치 소비의 변화 과정 연구

This study conducted text mining based on domestic news data to analyze changes in cosmetics consumption culture before and after the COVID-19 pandemic. The periods before the pandemic (January 20, 2019∼January 19, 2020) and after the pandemic (May 12, 2023∼May 11, 2024) were defined as comparison timeframes, and a total of 113,443 news articles collected through BigKinds, a news big data platform operated by the Korea Press Foundation, were used for analysis. The analytical methods applied were term frequency analysis, network centrality analysis (degree, closeness, and eigenvector centrality), and topic modeling (LDA). The results revealed that in the pre-pandemic period, keywords such as “celebrity,” “fan,” “shopping mall,” and “color” were central, reflecting symbolic and emotionally driven consumption, whereas in the post-pandemic period, functional and technology-centered keywords such as “COVID-19,” “pharmaceutical,” “artificial intelligence,” “formulation,” and “safety” newly emerged. The network analysis also showed a reorganization of highly central keyword clusters, and the topic modeling analysis confirmed a shift in consumer value perception-from external appearance to health, hygiene, ethics, and technology-based consumption. This study makes an academic contribution by empirically examining the transformation of the social meaning of cosmetics consumption through a comparative analysis of discourse structures before and after the pandemic based on unstructured text data. It is also expected to offer practical implications for designing post-pandemic consumer experiences and establishing brand strategies.