텍스트 마이닝을 통한 패션 기업의 ESG 정보 유형화

The aim of this study was to investigate fashion firms' environmental, social, and governance (ESG) information disclosure. A total of 25 fashion firms (e.g., Adidas, Burberry Group, Nike, Ralph Lauren Corp.) were selected, including eight luxury brands and eight athleisure brands. Thus, three groups were formed for analysis: the entire group (N = 25), luxury brands (N = 8), and athleisure brands (N = 8). Based on the ESG information disclosed on the firms' official web pages, 1128 valid words were extracted. The top keywords for each brand group were identified based on the frequency and term frequency-inverse document frequency (TF-IDF), and semantic network analysis and convergence of iterated correlations (CONCOR) analysis were performed. The results revealed that several keywords and clusters emerged with respect to unique attributes of the fashion industry, and they also revealed inconsistent ESG clusters according to brand type. The findings have significant academic and managerial implications.