基于网商评论数据的产品质量薄弱环节分析  

Weak-link Analysis of Products based on The Consumer Reviews

作　　者：孔雪峰 李东霖 杨军[1] 赵宇[1] Kong Xuefeng;Li Donglin;Yang Jun;Zhao Yu(School of Reliability and Systems Engineering,Beihang University,Beijing 100191,China;School of Computer Science and Engineering,Beihang University,Beijing 100191,China)

机构地区：[1]北京航空航天大学可靠性与系统工程学院,北京100191 [2]北京航空航天大学计算机科学与工程学院,北京100191

出　　处：《质量与可靠性》2021年第3期15-20,共6页Quality and Reliability

摘　　要：基于中文网商评论,提出了一种产品质量薄弱环节分析方法,为企业提供产品质量改进方向。首先,开展数据预处理剔除无用信息,并利用结巴(Jieba)对评论分词和词性标注;其次,基于词性路径模板提取评论特征;然后,基于情感分析抽取负面评论,并利用语义相似性开展主题聚类,确定评论主题;最后,针对评论主题进行意见归纳,确定产品薄弱环节。此外,以手机评论为应用案例,证明了所提出的产品质量薄弱环节分析方法的有效性。This paper develops a framework of weak-link analysis for products based on the Chinese consumer reviews,which is helpful in guiding the quality improvement of products.The consumer reviews are firstly pre-processed to remove the useless information,and the Jieba is applied for the word segmentation and part of speech(POS)tagging.Then,the review features are extracted according to the POS path templates.Next,negative comments are collected by the sentiment analysis,and the opinion topics of those reviews are obtained by clustering according the semantic similarity.Finally,the weak-links of products are provided through the summarization of opinion topics.A real case study of mobile phone reviews is given to show the effectiveness of the proposed method.

关 键 词：中文网商评论 薄弱环节分析 情感分析 主题聚类 意见归纳 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]