摘要
在电子商务环境下,赢得顾客满意是建立和保持长期客户关系的关键,因此需要建立有效的模型对顾客满意感进行评价和测量。本文以在线商品评论为数据来源,采取基于框架的情感语义分析方法,得出顾客满意感的评价属性、属性值及权重数据,与以往研究普遍采用的问卷方法相比,情感语义分析得到的信息更真实,数据处理效率更高;在情感分析的基础上采用VIKOR多属性决策法测量顾客满意感水平,该方法能够反映评价者的主观偏好,平衡考虑群体最大效用和个体遗憾,从而得出最优的评价方案。最后,通过一个真实数据的实验,验证了方法的可行性与有效性。
Under the environment of e-commerce, obtaining customer satisfaction is the key to maintain long-term customer relationships, thus it is necessary to establish an effective model for the evaluation and measurement of customer satisfaction. This paper, using online product reviews as data source, analyzed sentiment semantic based on frames and obtained the evaluation attributes, values and the weights. Compared to the questionnaire method previously used by most researchers, semantic analysis can get more real information and more efficient data processing. This paper uses the VIKOR multi-criteria decision making approach to measure customer satisfaction, which can reflect subjective preference and balance the maximum group utility and individual regret, and thus put forward the optimal evaluation. Finally, it validated the feasibility and effectiveness of the method by a real data experiment.
出处
《情报学报》
CSSCI
北大核心
2015年第10期1098-1110,共13页
Journal of the China Society for Scientific and Technical Information
基金
国家自然科学基金项目:基于FrameNet的中文评价词汇本体构建与观点挖掘研究(71403154)
教育部人文社会科学研究规划基金项目(12YJA740095)
山西省留学回国人员科技活动项目择优资助经费(2014058016)