The Research About The Dynamic Evolution Of Group Decisions Under The Network Environment

In recent years, the group decision theory has been studied widely. Experts have achieved fruitful achievements, including the concept and form of group decisions as well as assembled algorithm, etc. But unfortunately, there has not yet been a systematic and dynamic understanding corresponding to dynamic group decision in the current studies. And the effect on different characteristic groups has not been researched enough.This article focuses on the interaction of individuals in the decision making process. It also discusses the influence of different factors on the group opinion convergence. Opinion dynamics theory is used to analysis the basic ideas of the model and the principle of construction, combined with the theory of dynamic group decisions. At the same time, scale-free networks are built as a vehicle for individual interactions according to complex networks. On the basis of opinion dynamics theory and scale-free networks, the dynamic group decision model is built and the rules about individual interactions are set up.Based on the theoretical research, the Matlab tool is used to modify the traditional Deffuant interaction model with the hot issues under the network environment. By gathering and applying the real network data, it is observed that the parameters in the model have an influence on the group decision process. And when adding the individual opinion leaders, the results shows that opinion leaders worked as a guide in the evolution process of group decision. The research also shows that other individuals will quickly turn to this opinion in a short time when the support for some opinions has absolute advantages. Eventually the majority of the individuals will support this opinion. It is according with the characteristics of decision-making individuals in the real decisions.In this article, the theory of group decision has been expanded. The dynamic of individual preference adjustment in decisions has been confirmed. The performance of interaction decisions has been improved. In a word, this article has a theoretical and realistic significance in decision-making process.