Research On Multi-attribute Group Decision-making Method Considering Group Interaction In Probabilistic Language Environmen

The rapid development of social network platforms has increased the complexity of decision-making situations,and made people communicate and interact more frequently.The traditional multi-attribute group decision-making method has been difficult to meet the everchanging decision-making needs of the Internet era.On the one hand,due to the uncertainty of decision-makers’ cognition and the fuzziness of their expression,it is difficult to collect decision-makers’ evaluation information.On the other hand,the decision-maker is not an independent individual,and interaction actions among members in a group can have an influence on the decision outcome.In order to solve the above issues,this paper uses probabilistic linguistic term set(PLTS)to express the evaluation information of decisionmakers,and brings the interaction of decision-makers in the group into the decision situation.This paper puts forward the research of multi-attribute group decision-making method considering group interaction under the probabilistic linguistic environment.The research contents are as follows:(1)A multi-attribute group decision-making method based on probabilistic linguistic trust relationship is proposed.The probabilistic linguistic trust relationship is constructed to measure the interaction of decision-makers,and its nature is used to determine the weight of decisionmakers.In order to scientifically and effectively process the group evaluation information and obtain the best alternative,this paper successively proposes the probabilistic linguistic Choquet average(PLCA)aggregation operator and probabilistic linguistic Choquet power average(PLCPA)aggregation operator.The latter not only considers the interaction of attributes,but also reflects the opinions of most decision-makers in the group.Finally,the scientific rationality of the proposed method is explained through the case application and comparative analysis.(2)A multi-attribute group decision-making method based on quantum-like Bayesian network is presented.The interaction of decision-makers is measured by the interference effect in quantum theory,and a quantum-like Bayesian network is constructed to simulate the complex group decision-making process,and the parameters in the model are set.A new method for solving the interference term is proposed,by determining the values of three unknown parameters: conditional probability on the quantum-like Bayesian network,the weight of decision-makers,and the cosine value of the phase difference of decision-makers.Finally,a numerical example and method comparison are used to illustrate the effectiveness and superiority of the proposal.The method proposed in this paper not only conforms to the information expression habit of decision-makers,but also considers the interaction of members in the group.From the theoretical level,it further improves and enriches the theoretical research of multi-attribute group decision-making.On the application level,it can meet the ever-changing decisionmaking needs in the network era.