Research On The Prediction Model Of Social Network User Behavior

With the rapid increase in the number of people in online social networks,there are more and more messages spread in social networks.However,there are very few related studies on how information is transmitted in social networks and affect user behavior.In daily life and work,people's thoughts and behaviors are often influenced by friends and colleagues.For example,they are influenced by friends in their lives to watch a movie,and in the school,they are influenced by classmates to purchase a book.In the traditional model,many of them predict the behavior of a user unilaterally from the influence of a friend or the influence of a preference,and do not consider that the user is also affected by multiple factors at the same time,such as external influence.The problem of predicting user behavior based on social networks is based on the premise of social networks.It analyzes the user's historical behavior records and predicts whether the user will do something in the future.For example,a friend of a user recommends a movie and predicts the user.Whether it will be affected by friends to see the movie.The main work of this paper is to analyze the influencing factors of user behavior in social networks and to model the prediction of user behavior.First of all,this article addresses the issue of influencing factors of user behavior in social networks in order to reveal the mechanism of social events.This article describes three factors that have a significant impact on user behavior:social influence,external influence,and user preferences,and proposes the use of Poisson processes to model these three factors,respectively.Based on the above three factors,the causes of each event can be explained very accurately.Secondly,this paper aims at predicting the behavior of users in social networks,integrates and models the event sequences generated by the above three factors through the Poisson process,in order to enable the model to achieve an optimal effect on the prediction of user behavior.An algorithm to estimate the parameters of the model,the proposed model can not only explain how information is transmitted in social networks,but also can explain why a particular event occurred.