In recent day's terrorism poses a threat to homeland security. It's highly motivated by the “net-war” where the extremist are organized in a network structure. The knowledge discovery process in network analysis has been done using supervised and unsupervised techniques which helps us to differentiate overt and covert nodes in a social network. The major problem faced is to automatically identify the key player who can maximally influence other nodes in a large relational covert network. The existing centrality based and graph theoretic approach are more concerned about the network structure rather than the node attributes. An unsupervised framework is used to describe the nodes and links in the form of a directional semantic graph where each node is related with more than one relationship with others. The behaviors of nodes are analyzed based on the semantic profile generated. The semantic profile is described as a collection of condensed paths generated through variable relaxation approach. These condensed paths are called as path types and each of them have a unique format. To identify the key player the outlier analysis is done in the profile and the highly communicating node is concluded to be the most influential node of the network.