摘要
本文利用文本挖掘方法分析了智能手环用户评论中的热点主题,以了解用户对不同品牌智能手环关注点差异以及用户关注主题随时间的变化。首先,利用LDA模型对文本进行主题分类,了解不同品牌用户关注主题的排序;然后对不同时段各品牌的评论信息进行分析,了解不同用户群体对产品功能关注度的变化;最后分析用户对不同品牌手环整体满意度的差异,并结合评论文本信息分析用户使用时长与满意度之间的关系。本研究发现对于所有品牌,用户均关注产品功能和外观,以及在使用中发现的产品问题和感受,但是不同品牌关注主题排序各有差异。就用户满意度而言,随着使用时间的增长,用户对小米手环的满意度日渐降低,而Fitbit和华为手环的用户满意度与使用时间呈正U形曲线。
This paper analyzes hot topics in online reviews of smart bracelet by using text mining tech- nique to understand the differences in users＇ attention towards different smart bracelet brands and the time- varying change of their focus. At first, we have a topic classification using LDA model by analyzing pre-pro- cessed text to know users＇ focused topics of different brands. Secondly, we analyze online reviews in differ- ent time stages to understand different user groups＇ time-varying attitudes to product functions. At last, this paper investigates the difference of customers＇ satisfaction towards different smart bracelet brands, and ex- plores the relationship between customer satisfaction and using time. We find that users of all brands care a- bout product function, outlook, and their feeling about using the product. However, the order of hot topics of different brands varies. As for the customer satisfaction, the satisfaction degree of Xiaomi users has been decreasing with the increase of using time, while the relationship between customer satisfaction and using time of Huawei and Fitbit is a positive U-shaped curve.
作者
吴江
周露莎
刘冠君
贺超城
Wu Jiang Zhou Lusha Liu Guanjun He Chaocheng(School of Information Management, Wuhan University, Wuhan, 430072 The Center for Electronic Commerce Research and Development at Wuhan University,Wuhan,430072)
出处
《信息资源管理学报》
CSSCI
2017年第3期24-33,共10页
Journal of Information Resources Management
基金
国家自然科学基金项目"内容关系互动下的在线医疗社区用户行为演化研究"(71573197)和"物联网环境下组织体系架构建模
行为分析与优化设计"(71531009)的研究成果之一
作者简介
吴江（通讯作者），男，教授，博士。研究方向为社会网络计算、网络信息计量、在线医疗社区，Email：jiangw@whuedu．cn
周露莎，女，硕士研究生，研究方向为在线医疗社区；
刘冠君，女，本科生，研究方向为可穿戴设备研究：
贺超城，男,硕士研究生，研究方向为社会仿真，数据挖掘与数据分析。
