基于产品性能词典的评论数据驱动客户需求模型研究  

Review Data Driven Customer Need Model Research Based on Product Performance Lexicon

作　　者：张文旭 肖人彬[1] 林文广 ZHANG Wenxu;XIAO Renbin;LIN Wenguang(School of Artificial Intelligence and Automation,Huazhong University of Science and Technology,Wuhan,430074;School of Mechanical and Automotive Engineering,Xiamen University of Technology,Xiamen,Fujian,361024)

机构地区：[1]华中科技大学人工智能与自动化学院,武汉430074 [2]厦门理工学院机械与汽车工程学院,厦门361024

出　　处：《中国机械工程》2020年第15期1866-1876,共11页China Mechanical Engineering

基　　金：国家自然科学基金资助项目(51875220);福建省自然科学基金资助项目(FJ2017J0102)。

摘　　要：利用数据驱动手段快速准确地从客户行为数据中获取客户需求信息是产品开发中亟待解决的问题,为此对客户需求模型进行研究,提出了评论数据驱动下客户需求建模的方法。首先,针对基于主题聚类的评论分析中主题数目不易确定的问题,提出了产品性能词典,用以确定聚类主题,对TF-IDF方法进行了改进,用于词典创建,并基于产品性能词典完成了客户需求挖掘。其次,针对现有产品评论研究缺乏需求分析过程的问题,提出基于产品评论数据的Kano分析方法,在此基础上引入物元表示法对客户需求模型进行量化表达。最后通过实例研究验证了所提方法的可行性与有效性。It was an urgent problem in product development to quickly and accurately obtain customer need informations from customer behavior data by means of data-driven method.Therefore,the customer need model was studied,and the customer need modeling method driven by review data was proposed.Firstly,aiming at the problem that the number of topics was not easy to determine in the review analysis based on topic clustering,a product performance lexicon was put forward to determine the clustering topics.The TF-IDF method was improved to be used for lexicon creation,and the customer need mining was completed based on the product performance lexicon.Secondly,in view of the lack of customer need analysis processes in current product review research,the product review data based Kano analysis method was proposed.On the basis of above metioned,matter-element theory was introduced to quantitatively express the customer need model.At last,the feasibility and effectiveness of the proposed methods were verified by a case study.

关 键 词：客户需求挖掘 客户需求分析 产品性能词典 KANO模型 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP391[自动化与计算机技术—控制科学与工程]