Clustering method based on centrality metrics for social network analysis

Abstract

The significance of a node in a social network is quantified through its centrality metrics, such as degree, closeness, and betweenness. However, many methods demonstrating the relevance of a node in the network have been proposed in the literature. In this digital smart world, the evolution of social networks occurs in various different directions at an unprecedented speed. A network evolution mechanism that provides the state of each node and its changes from its inception to its extinction over time will help in understanding its behavior. Often the strategy behind evolution is unknown and would not be reproduced in its totality. However, it is essential to understand behavior of the network as this can greatly facilitate its management before it becomes uncontrollable. A heuristics-based cluster method is proposed in this paper which combines centrality metrics and categorizes the entire network.