| With the rapid development of the Internet,new forms such as live streaming and short videos have changed consumers’ consumption scenes.Social media such as the Little Red Book has become an important source of information affecting consumers’ purchasing decisions.Meanwhile,based on consumers’ information needs,a large number of opinion leaders based on product experience and content creation have emerged on social media platforms,attracting the participation of many consumers.These opinion leaders usually have superior status,social prestige and more experience and information about products or services,thus influencing consumers’ buying decisions and behaviors.The buying decision-making process of consumers usually includes information exposure,attention,interest,reception,memory,acceptance,decision-making,etc.By taking advantage of the purchasing decision-making process of consumers,social network opinion leaders realize the dynamic guidance of consumers’ purchasing decisions.Exploring the dynamic guidance process of opinion leaders can clarify the important role of opinion leaders and help enterprises to enhance the effect of network marketing and product promotion,which has important practical significance.Currently,the purchase decision of social network consumers mainly studies the static process from the influencing factors of consumer purchase decision to purchase behavior.But in fact,consumers’ purchase decision process needs to go through receiving stimuli,obtaining information to persuasion,and then forming decisions and behaviors,which is a dynamic process.Therefore,researchers rely on the theory of opinion dynamics to build the bounded trust model to explore the dynamic process of propagating ideas.Research on the Hegselmann-Krause(HK)model of opinion dynamics pays little attention to the degree of closeness and similarity of opinions.This paper considers the influence of closeness and similarity of opinions on the dynamic guidance process of opinion leaders on consumers’ purchase decisions in social networks,which is an important supplement to the research on social networks.Based on the theory of opinion dynamics,an improved bounded confidence HK model was built by introducing opinion similarity and closeness.Using Python simulations,the dynamic influence of social media opinion leaders on consumer buying decisions was explored.Considering the relationship between consumers and consumers in social networks,the paper sets the opinion updating rules of positive opinion leaders,negative opinion leaders and opinion followers in the network for modeling,and analyzes the dynamic guidance process of opinion leaders to consumers through simulation experiments.Experimental results show that when the point of opinion similarity and closeness are high,the degree of opinion concentration of followers is higher.When the number of positive opinion leaders is slightly higher than that of negative opinion leaders,the influence of the positive opinion leaders is the largest,but the influence will not increase indefinitely.As the confidence level of followers increases,the influence of opinion leaders on followers increases,and followers ultimately achieve consensus.The more that followers trust positive opinion leaders,the more important the role of positive opinion leaders.Lastly,the efficiency of the improved model is demonstrated through comparative experiments. |