Large Group Consensus Reaching Method With Trust Network And Its Application

As decision making processes become more complex,decision makers face greater challenges and a few decision makers are unable to make more scientific decisions.On the one hand,solving a major problem requires combining the wisdom of experts in different fields to get a persuasive decision result,which makes the number of decision makers become more and more numerous and coordinating decision makers with different preferences to get consistent results becomes difficult.On the other hand,the impact of modern information technology on human beings has penetrated into many aspects,such as Taobao has changed people’s shopping habits,We Chat and Tiktok have changed people’s communication habits,and Di Di has changed people’s travel habits.Information technology has deepened communication among people,and decision makers are connected to each other in social networks and have certain relationships with each other,which are important factors influencing decision making.In the social network environment,solving the large group decision making problem faces the challenge of preference differences and complex relationships among decision makers in different domains.In order to better solve the large group decision making problem in practice,this thesis integrates social network analysis methods and decision making methods to study the influence of trust relationships among decision makers on the decision making process.The main work of the thesis focuses on the following five aspects.(1)In large group decision making problems,due to the large number of decision makers,the decision makers cannot give the trust estimation value of all decision makers in the group,so there is a problem of missing trust relationship among decision makers.Based on the propagation characteristics and combinability of trust,this thesis constructs individual trust networks and distrust networks of decision makers through large group trust networks,so as to obtain individual trusted reference values and distrusted reference values of decision makers,and collect these two reference values to obtain individual trust levels of decision makers.The individual trust level can help decision makers to make an objective trust estimation.(2)The existence of trust relationships is dynamic in nature due to the fact that they are not static,but the trust values change dynamically as the decision makers communicate with each other.In the decision making process,the initial trust relationship between decision makers is obtained based on a priori knowledge,however,as the decision making process evolves,decision makers gain new experiences and update their relationship with each other accordingly.Trust relationships are event-sensitive in nature,which means that new experiences have a greater impact on trust relationships than old experiences.Since trust relationships not only influence clustering process,but also play an important role in adjusting decision makers’ opinions,the updating of trust relationships is also an important factor affecting decision making.(3)In social network large group decision problems,decision makers are not independent of each other but are connected through trust relationships,so that during the clustering process,decision makers not only want to be in a subgroup with decision makers who have similar preferences to their own,but also with decision makers whom they trust.Thus,both trust relationships and preferences are important indicators of clustering.This thesis presents a grey clustering algorithm that incorporates preference information and trust relationships.(4)Decision making is a dynamic process,accompanied by the adjustment of decision-makers’ preferences and the updating of trust relationships.This article introduces trust relationships into the consensus reaching process and proposes a two-stage consensus reaching method.In the first stage,based on the matching degree between the decision maker’s preferences and the trust relationship,and whether the decision-maker is in the same subgroup,it is determined that the decision-maker updates the trust relationship or adjusts their preferences to obtain a more realistic trust relationship and achieve the goal of not damaging the cohesion of the subgroup.In the second stage,considering the decision maker’s preference adjustment willingness and subgroup constraints,four adjustment scenarios are analyzed,and adjustment rules are given in turn.(5)A practical example of power battery supplier selection is applied to illustrate the effectiveness and feasibility of the proposed method.Compared with the existing methods,the advantages and effectiveness of the method are verified.