电商平台服装畅销影响因素研究——基于Logistic回归与文本挖掘

Research on the Influencing Factors of Clothing Sales on E-Commerce Platforms—Based on Logistic Regression and Text Mining

摘要: 为探究影响电商服装畅销的因素，本研究结合Logistic回归与文本挖掘方法，分析了淘宝平台8800款商品数据及2400条评论。Logistic回归分析表明，“七天退换”、“正品保障”、“天猫认证”和“有视频”显著促进服装畅销，而“性别”(男装)、“价格”及“折扣”则抑制服装畅销；文本挖掘进一步验证了产品质量与体验的核心地位，揭示了性别差异：女性消费者更关注服装细节与款式，男性则更看重功能实用性与购物效率。本研究为商家提供了基于实证的运营优化建议。

Abstract: To investigate the factors influencing the sales of clothing on e-commerce platforms, this study employed logistic regression and text mining methods to analyze data from 8800 product listings and 2400 customer reviews on Taobao. The Logistic regression analysis revealed that “7-day return and exchange”, “authenticity guarantee”, “Tmall certification” and “product video” significantly promoted sales, whereas factors such as “gender” (men’s clothing), “price” and “discount” had inhibitory effects. Text mining further validated the core importance of product quality and user experience, highlighting gender-specific differences: female consumers placed greater emphasis on clothing details and aesthetics, while male consumers prioritized functional utility and shopping efficiency. This study provides empirically supported operational recommendations for e-commerce merchants.