Influence Maximization In Weibo Based On Community Structure And Node Attributes

Marketing workers utilizes social network to make a small number of people use enterprise's products or accept its brand concepts,leading to the word of mouth.Finally,these enterprise's products or brand concepts spread rapidly in the social network,achieving promoting products or brand.How to find out those initial a few users to achieve the widest spreading scope has been becoming a research focus in recent years,namely Influence Maximization.Since Richardson introduced the influence maximization into the analysis area of social network,many scholars have made great efforts to this field and gotten plenty of good achievements.This paper is going to study the Influence Maximization though the following ways:First,starting with the community structure in the network,people can divide the network into several communities through the effective community discovery algorithm.This paper will choose the CNM algorithm,using the module degrees to measure whether the node should join a community.Second,choose the communities which have appropriate number of nodes and solve the problem that in which community to find the Nth node through the dynamic programming algorithm.When finding the community through dynamic programming method,people can make the method more practically significant by joining node properties –the degree and tweeting.Finally,through the basic influence maximization algorithm—KKT algorithm,scholars can find the Nth node and assemble all the nodes,namely the initial node set.This paper boosts the pretty flexibility and can achieve very good results.