Research On The Influence Of Social Network Personalization On Communication

With the rise and development of social networks,the issue of maximizing impact has become a hot spot in social networks.However,the traditional impact maximization problem can no longer meet the diversified needs of current application scenarios.Therefore,the maximization of personalization influence has become popular as a branch of social network impact maximization.The purpose is to target a specific social network user and mine the initial impact communication user set that maximizes its impact.In this paper,from the two aspects of propagation model and propagation mode,this paper explores the problem of maximizing individualized influence,and proposes the influence maximization algorithm based on multi-cascade model and the influence maximization algorithm based on heat diffusion model.(1)Personalization influence maximization based on multiple cascading models.The multi-cascade model is an extension of the traditional independent cascade model.In the traditional IC model,the state of users is only activated or inactive,which cannot well simulate the real propagation process.In a multi-cascade model,the user state is the sum of the number of activation times.In this paper,we study the maximization of individuation effect under multiple cascade model.At a given time of promotion time,a set of target users,and a network in which each node can be activated multiple times by its neighbors,and our goal is to identify the k most influential seeds to maximize limit the total frequency of activations received by these target users.In this paper,a multi-objective impact maximization algorithm based on multi-cascade model is proposed.The frequency of the affected users is used to measure the impact intensity on the target users,and the candidate users are clustered.The cluster center is used as the seed node to expand information dissemination.The experimental results show that the multi-cascade model proposed in this paper can achieve a wider propagation range in less time.(2)Personalization influence maximization based on heat diffusion model.Since the existing methods are mostly based on the IC model and the LT model,theprediction of the influence of the two models on the nodes depends on the Monte Carlo simulation.In order to avoid Monte Carlo simulation time loss and more practical life,this paper introduces the heat diffusion model into the problem of maximizing individualized influence,and uses the heat diffusion process to simulate the propagation process of information influence.Thermal energy is used to measure the impact on the target users,and the candidate users are clustered,and the cluster center is used as the seed node to spread information to maximize the impact on specific users.This paper proposes an optimization method based on breadth-first traversal to reduce the size of candidate sets.The experimental results show that the proposed propagation mechanism based on the heat diffusion model can achieve a wider influence propagation range in less time.