基于K-means与Word2vec的哺乳文胸评论主题挖掘研究  

Research on Topic Mining of Breastfeeding Bra Comments Based on K-means and Word2vec

作　　者：刘妍 刘驰[1] LIU Yan;LIU Chi(Apparel&Art Design College,Xi'an Polytechnic University,Xi'an 710048,China)

机构地区：[1]西安工程大学服装与艺术设计学院,西安710048

出　　处：《人类工效学》2024年第2期40-45,共6页Chinese Journal of Ergonomics

基　　金：国家自然科学基金项目(72171188)。

摘　　要：目的为了了解消费者在网络平台购买哺乳文胸时的关注侧重点,文章从在线评论中抽取有效关键词构建哺乳文胸主题,并通过计算主题的重要程度协助商家了解消费者关注重点方向。方法选用TF-IDF关键词抽取算法,结合K-means和Word2vec进行语义聚类、主题识别、主题词挖掘及主题重要度计算。结果哺乳文胸评论文本聚类后的主题重要程度排名是:产品品质(45.47%)、产品外观(35.83%)、产品服务(18.79%)。结论通过该方法能够有效的识别和构建哺乳文胸主题及主题词,同时,通过主题的重要程度,能够了解消费者对于网络平台购买哺乳文胸时关注的重点方向,为哺乳内衣企业进行产品改善及生产等提供理论参考。Objective In order to understand the focus of consumers when purchasing breastfeeding bras on online platforms,the article constructs breastfeeding bra themes by extracting effective keywords from online reviews,and assists merchants in understanding the direction of consumers'focus by calculating the importance of the themes.Methods The TF-IDF keyword extraction algorithm is selected,combined with K-means and Word2vec for semantic clustering,topic identification,topic word mining and topic importance calculation.Results The results showed that the ranking of topic importance after clustering breast feeding bra review texts was:product quality(45.47%),product appearance(35.83%),and product service(18.79%).Conclusion The results confirm that the method can effectively identify and construct the breastfeeding bra theme and theme words,and at the same time,through the importance of the theme,it can understand the key direction of consumers'concern for the purchase of breastfeeding bra on the online platform,which provides theoretical references for breastfeeding underwear enterprises to improve their products and production and soon.

关 键 词：服装工程 文本聚类分析 哺乳文胸 在线评论 K-MEANS Word2vec 主题挖掘 主题重要程度 文献计量分析 

分 类 号：TS941.1[轻工技术与工程—服装设计与工程]