This paper takes numerical online evaluation information as the research object,considering the situation that the weight of the commodity attribute is unknown and the relevance of the commodity attribute,and considering the psychological behavior of the decision maker,the problem of commodity ranking based on numerical online evaluation information is studied.Considering that the attribute weight is unknown and the attribute relevance degree,the structural hypothesis is verified by the structural equation model,and the attribute and its weight are determined according to the proportion of the influencing effect of the attributes.On this basis,considering the psychological behavior of the decision maker,two different kinds of commodity ranking method based on numerical online evaluation information are proposed.One is based on prospect theory.Firstly,the expected value of the decision maker for different attributes is considered,and the expected value of the attribute is used as a reference point to calculate the value function.Then,the probability weight function is calculated.Finally,the foreground value is calculated,and all items are sorted according to the comprehensive foreground value.The other is based on the prospective random dominance criterion.Firstly,according to the reference point of the decision maker,the decision matrix is transformed into the profit and loss matrix,and the probability distribution of the profit and loss matrix is calculated.Then,based on the prospective random dominance criterion,the foreground stochastic dominant matrix is constructed.Finally,the PROMETHEE-? method is used to sort the items.The example results show that both methods can solve the problem of commodity ranking based on numerical online evaluation information,which verifies the feasibility of foreground theory and prospective random dominance criterion in solving commodity ranking problem. |