Group Decision-making Model Considering Interference Effect From The Perspective Of Social Networ

The core of group decision-making theory is a process of gathering collective wisdom.And the development of social networks has strengthened the relevance between people,which requires the application of social psychology to overturn the independence hypothesis of the decision-making subject in traditional group decision-making,thus providing new inspiration for the research of group decision-making theory.In this paper,the quasi-quantum Bayesian network in quantum probability is used to fit the interference generated when calculating the trust information and feedback information.Combined with fuzzy measures,an incomplete preference consensus reaching method based on interaction is proposed to provide new solution and decision support for emergency decision-making,voter voting,resource allocation and other decision-making issues in the social network environment.The main research contents of this paper are as follows:(1)A trust measurement method based on the interference effect between trust sources.Starting from the social relationship and background of decision-makers,this paper classifies trust sources and proposes "familiarity" and "professionalism" as indicators of trust sources,which avoids the redundancy of trust sources effectively.The interference between trust sources is described based on quantum probability theory.At the same time,a new trust propagation operator is proposed to deal with the indirect trust relationship.(2)Group decision-making model based on trust interaction.The trust score is converted into the interaction index,which is used as a supplementary item of the common weight to reflect the synergy or redundancy effect between decision-makers due to "interaction".Combined with Choquet integral,the calculation method of the utility value of scheme is designed to reflect the impact of the interaction between decision-makers on the enhancement or weakening of the utility value,so as to obtain a more accurate ranking result of the scheme.(3)Bidirectional feedback model considering interference effect.The quasi-quantum Bayesian network is constructed to describe the game process between the consistent group and the inconsistent group.The probability of modifying the preference is expressed by quantum probability,and the interference term is determined as the basis for discussing the interference effect when the group preference is adjusted.Combining the interference term with the feedback model,a bidirectional model driven by the minimum feedback control parameter(i.e.the minimum adjustment step)is proposed to ensure that the consensus level can be met with the minimum adjustment.