Logistic回归与随机森林模型在新中式服装消费行为影响因素中的应用
The Application of Logistic Regression and Random Forest Model in Influencing Factors of New Chinese-Style Clothing Consumption Behavior

Dior“文化挪用”马面裙事件引发民众热议，以“马面裙”为代表的新中式服装走进大众视野。新中式服装实现了传统文化与现代时尚的完美融合，其产业发展在经济、文化等不同维度上均具有重要价值，因而了解民众对新中式服装的情感倾向与关注点，深入分析影响民众关于新中式服装购买意愿的影响因素对新中式服装产业的未来发展及传统文化传承具有深远意义。团队通过网络爬取进行文本分析，为问卷设计提供理论支持。基于描述性统计分析构建消费者画像，综合应用二值Logistic回归与随机森林模型确定影响消费意愿的主要因素，从消费者因素、产品因素和推广途径三个方面明确新中式服装的市场方向，以此提出有价值的结论与建议。结果表明，性别、年龄、学历、收入、地区作用的消费者自身因素是影响新中式服装购买意愿的最主要因素。
Dior’s “cultural appropriation” incident involving horse-face skirts has sparked public discussion, and new Chinese-style clothing represented by the horse-face skirts has entered the public eye. This style perfectly blends traditional culture with modern fashion, and its industrial development has important value in different dimensions such as economy and culture. Therefore, understanding the emotional tendencies and concerns of the public towards the new Chinese-style clothing, and conducting a deep analysis of the factors influencing the public’s willingness to purchase such clothing have profound significance for the future development of the new Chinese-style clothing industry and the inheritance of traditional culture. The team conducted text analysis through web crawling to provide theoretical support for questionnaire design. Based on descriptive statistical analysis, a consumer portrait is constructed, and binary logistic regression and random forest models are comprehensively applied to determine the main factors affecting consumers’ willingness. Clarifying the market direction of new Chinese-style clothing from three aspects: consumer factors, product attributes, and promotional channels, in order to provide valuable conclusions and suggestions. The results indicate that consumer factors such as gender, age, education, income, and regional influence are the most significant factors affecting the willingness to purchase new Chinese-style clothing.