Agent-Based User Purchase Behavior Modeling In Social Network And Experimental Analysis

The modeling of social network users' purchase behavior aims at simulating people's buying behavior to the greatest extent,and analyzing their inner working mechanism,so as to grasp the general laws of consumers' buying behavior.However,due to the heterogeneity of individuals and the diversity of purchasing influence factors have brought difficulties to the development of this work.Most of the previous studies in this area are based on statistical methods.From a macro perspective,users' purchase behaviors are analyzed from the top down and there is no in-depth exploration of the user's internal decision-making mechanisms.In recent years,the rapid development of computer technology has provided new ideas for researching users' buying behavior.This paper adopts an agent-based method,analyzes the mechanism of action between different influencing factors,presents the user's decision model,and studies the user's purchasing behavior from a microscopic perspective.First,starting from the specific application scenarios,this thesis extract the technology demand of user purchase behavior modeling.Secondly,summarizing three main factors that influence consumers' purchase decisions: social network,network reviews and advertisements.The impact strength of different factors on consumer decision-making and the evolution mechanism of influence intensity are given.On this basis,a consumer decision association network is constructed.The whole purchase process is divided into four stages: demand perception,information search,purchase decision and experience feedback,and the dual system decision based on neuroscience is introduced.At the same time,the purchase fluctuation of users is considered,and a model of user purchase decision based on psychological process is put forward.Through the use of NetLogo software for experimental analysis,and compared with the classic expected utility model(EU model)and the subjective expected utility model(SEU model)to verify the validity and accuracy of the model.The experimental results show that the decision making model based on mental process can improve the accuracy and efficiency of the prediction.From the consumer's own point of view,the individual's confidence in advertising,the praise rate of the network reviews,and the feedback from the interpersonal relationship will affect the individual's purchase decision.At the same time,it is found that this model has a better prediction effect in the case of high consumer involvement.