The Role of Consumers' Subjective Visual Preference in Fashion Product Acquisition

This study investigates the role of consumers' subjective visual preferences in fashion product acquisition, focusing on how visual similarity and preference variation influence purchasing behavior. Using a dataset from a fashion enterprise, which includes over 1.7 million records spanning three years, the research employs deep learning techniques, specifically the Gram matrix and VGG19 model, to analyze the visual features of fashion products. The study reveals that consumers tend to purchase products that are visually distinct from their previous acquisitions, and those with diverse visual preferences demonstrate higher purchase intentions. The findings contribute to the literature by integrating visual attributes into consumer behavior analysis, offering valuable insights for marketers to enhance sales strategies and foster customer loyalty.