온라인 리뷰 텍스트마이닝을 통한 여성 소비자의 긍정 감정을 유발하는 메이크업 제품 속성 분석

Abstract

The purpose of our study is to identify the attributes of makeup products that elicit positive emotions in female consumers by integrating text mining, semantic network analysis, sentiment analysis, and CONCOR analysis. We collected online reviews from major Korean e-commerce and community platforms over the past two years and preprocessed them using the Textom system. Fifty key attributes were identified and extracted for subsequent analysis. The highlighted exploratory frequency analysis frequently mentioned terms, such as blockbuster, smudge- proof, and oil-free, while network centrality revealed subtle scent, non-sticky, and plumping as influential hubs. Positive emotions were predominantly prevalent while negative emotions were marginal when profiling sentiments. The CONCOR analysis clustered these attributes into five groups: performance-oriented, comfort-focused, visual- radiance, emotional-value, and mildness-trust. This demonstrated that durability, multisensory comfort, and safety assurance collectively underpin the positive emotions. Theoretically, our study contributes an integrated analytical framework to understand the relationships among textual emotions, relational prominence, and structural equivalence. On a practical level, it suggests modular product strategies that emphasize long-lasting, non-sticky, and subtly scented formulations combined with the implementation of real-time monitoring systems to detect safety-related cues in consumer reviews. Future research could benefit from expanding data sources and performing longitudinal analyses to trace evolving consumer sentiments.