Abstract: A probabilistic linguistic multi-attribute variable weighted decision making method is proposed by using a large amount of online evaluation information. Firstly, online evaluation information with respect to each attribute is described as probabilistic linguistic term, the decision matrix about probabilistic linguistic is constructed, and the score function of the interval probabilistic linguistic term is redefined. Secondly, by variable weighted analysis method to deal with decision matrix, we can get the attribute weights of different alternatives. Then, based on regret theory, the alternatives' comprehensive perceived utilities can be calculated and ranked accordingly. Finally, with the online evaluation information provided by the website ‘autohome.com.cn', the feasibility and rationality of the proposed method in this paper are proved through an analysis of a car selection example.

Key words: online evaluation information, probabilistic linguistic term, multi-attribute decision making, Regret theory, variable weighted method

摘要： 针对电子商务网站中涌现的大量在线评价信息,本文提出一种基于在线评价信息的概率语言多属性变权决策方法,指导消费者的购买决策。在本文方法中,首先利用概率语言描述在线评价信息,构建概率语言决策矩阵,并重新定义了区间概率语言的得分函数;其次,利用变权方法处理决策矩阵,得出不同方案的各属性权重;再次,依据得出的属性权重,基于后悔理论,考虑决策者风险规避系数,求出各方案的综合感知效用值,并据此排序。最后,利用汽车之家网站提供的顾客在线评价信息,以一个汽车选择的实例说明了本文所提方法的合理性与实用性。

关键词: 在线评价信息, 概率语言, 多属性决策, 后悔理论, 变权方法