管理评论 ›› 2025, Vol. 37 ›› Issue (2): 163-174.

• 市场营销 • 上一篇    

文化产品评论主题对有用性影响的跨文化研究

肖娴1, 邱凌云2, 庞隽3   

  1. 1. 中国科学院文献情报中心, 北京 100190;
    2. 北京大学光华管理学院, 北京 100871;
    3. 中国人民大学商学院, 北京 100872
  • 收稿日期:2022-08-23 发布日期:2025-03-06
  • 作者简介:肖娴,中国科学院文献情报中心馆员,博士;邱凌云(通讯作者),北京大学光华管理学院副教授,博士生导师,博士;庞隽,中国人民大学商学院副教授,博士生导师,博士。
  • 基金资助:
    国家自然科学基金面上项目(72272004;72072179)。

A Cross-cultural Study on the Influence of Textual Topics on Review Helpfulness of Cultural Products

Xiao Xian1, Qiu Lingyun2, Pang Jun3   

  1. 1. National Science Library, Chinese Academy of Sciences, Beijing 100190;
    2. Guanghua School of Management, Peking University, Beijing 100871;
    3. School of Business, Renmin University of China, Beijing 100872
  • Received:2022-08-23 Published:2025-03-06

摘要: 本研究基于跨文化心理学的相关理论,使用数据驱动的主题模型,比较了图书在线评论中不同内容主题在对评论有用性影响上的中美差异。通过分析来自亚马逊中国和亚马逊美国的50270条评论,研究发现,与“建议与心得”类或“对产品的双边或中性评价”等主题相关的评论内容对中国消费者比对美国消费者更有用,而与“对产品的单边评价”等主题相关的内容对有用性的影响则不存在跨文化差异。这些发现不仅丰富了关于在线评论的跨文化比较和评论有用性影响因素的文献,还有助于跨境零售企业为不同文化背景下的消费者提供更有针对性的产品评论。

关键词: 在线评论, 跨文化差异, 评论有用性, 文本挖掘, 主题模型

Abstract: Based on theories in cross-cultural psychology, this study employs a data-driven topic model to compare how different content topics in online review texts of books affect the evaluation of review helpfulness by Chinese and American consumers. Using the Latent Dirichlet Allocation (LDA) topic model, we extract eight general topics about books from 50,270 reviews posted on the websites of Amazon China and Amazon USA. By comparing the impact of various topics on review helpfulness between Chinese and American reviews, we find that review content related to Reading Experience/Purchasing Advice and bilateral or neutral Attribute Evaluation is more useful to Chinese consumers than to U.S. consumers. By contrast, Chinese and American consumers do not differ in their assessment of the helpfulness of review content related to informational topics. These findings enrich the literature on cross-cultural comparisons of online reviews and factors influencing review usefulness. They could also help cross-border e-commerce retailers optimize the design of their online review systems by providing more targeted product reviews for consumers of different cultural background.

Key words: online reviews, cross-cultural differences, review helpfulness, text mining, topic modelling