Influence Maximization Algorithm Based On Incremental Gaussian Diffusion Of Nodes

With the development of technology,the continuous advancement of the Internet has promoted the rapid development of social networks.Nowadays,social networks have penetrated into all aspects of people's lives,but the large amount of data brought by social networks has also generated new social research issues and the impact of social networks Maximizing is one of them.Impact maximization is currently one of the hottest issues in social networks,but there are still difficult issues such as source monitoring of public opinion,selection of marketing agents,source monitoring of water quality,etc.,which need to be improved in terms of accuracy and time cost.The research focus of this paper is on the processing and modeling of node data attributes,and then by simulating the propagation of the influence of each node in the social network to maximize the location of the most influential nodes in combination with IC models.This method mainly defines the attractive force existing between nodes for incremental accumulation,and aims to use the accumulated influence to calculate the influence probability between nodes,so as to improve the accuracy and speed of finding the main nodes.The main breakthroughs in this paper are as follows:First of all,in researching the impact of social network maximization,the attribute relationships between nodes and nodes overlap,and there are repeated calculation problems in the calculation process.This paper will mark overlapping nodes by dividing the overlapping communities to improve the accuracy of searching nodes.Then,two attributes are defined in the initially processed network nodes: The nodes themselves paper;The cumulative propagation power between the nodes.This is proposes the Influence Maximization Based on Node Attraction Model to Improve the accuracy of searching nodes as a whole through two attributes.Furthermore,gaussian diffusion model is introduced to improve the time cost of experimental simulationFinally,this paper is using two different data sets for experiments.The results show that the algorithm model proposed in this paper can be improved in terms of accuracy and speed.