Locating Key Actors in Social Networks Using Bayes' Posterior Probability Framework

Typical analytical measures in graph theory like degree centrality, betweenness and closeness centralities are very common and have long history of their successful use. However, modeling of covert, terrorist or criminal networks through social graph dose not really provide the hierarchical structure of such networks because these networks are composed of leaders and followers. It is possible mathematically, for some graphs to estimate the probability that the removal of a certain number of nodes would split the networks into may be non functional network. In this research we investigate and analyze a social network using Bayes probability theory model to calculate entropy of each node present in the network to high light the important actors in the network. This is accomplished by observing the amount of entropy change computed by successively removing each node in the network.