소셜 빅데이터 분석을 활용한 디저트 카페 인식 비교 연구

Abstract

This study aims to identify the structural characteristics of consumer perceptions and explore future industrial implications by collecting unstructured big data related to dessert cafes from social media platforms and analyzing it through systematic text-mining procedures. The analysis was conducted by dividing the study period into two phases—the COVID-19 outbreak period and the post-pandemic recovery period—to examine temporal differences in consumer perception. The analysis employed frequency analysis, centrality analysis, and CONCOR analysis to conduct comparative assessments. The findings reveal that, during the outbreak period, keywords such as “nowadays,” “beverage,” and “atmosphere” exhibited high frequency, reflecting consumers’ focus on nearby café visits and aesthetic experiences. In contrast, in the post-pandemic period, keywords such as “atmosphere,” “Seoul,” and “emotion” were dominant, indicating a shift toward experiential and travel-oriented consumption patterns. The degree centrality analysis demonstrated that keywords associated with urban and local café visits showed high centrality in both periods.