Introduction of the Entropy-Based Method for Finding Influential Nodes in Information Dissemination on Online Social Networks

"In cognitive cyber-attacks, information dissemination analysis in online social networks is a very important issue. One of the main branches of information dissemination analysis is finding influential nodes which this issue is also arisen in viral marketing as finding the influential individuals. In this paper, while introducing and calculating two types of important nodes in information dissemination (reference and active nodes), a method for identifying these two important types of nodes in the dissemination of information in online social networks is presented and implemented based on the entropy theory. The proposed method in this paper is based on the evaluation of the entropy of the online social network graph generated from dissemination of information by removing the set of the most influential nodes measured on the basis of the nodal-degree and the entropy of the nodes. The experiments of this paper show that the proposed algorithm is more capable of identifying the set of influential nodes than the previous methods, in a way that the remaining set of nodes will have an adjustable homogeneity in influence measure and also presents a measure for determining the number of influencer nodes.