Efficient Extraction of High-Betweenness Vertices from Heterogeneous Networks

摘要: Centrality measures are crucial in quantifying the roles and positions of vertices complex network analysis. An important popular measure is betweenness centrality, which computed based on number shortest paths that fall on. However, computationally expensive to derive, resulting much research efficient computation techniques. We note many applications, it set with high key interest their rankings rather than exact values usually adequate for analysts work with. Hence, we have developed a novel algorithm efficiently returns highest betweenness. The convergence criterion our membership stability high-betweenness set. Through experiments various artificial real-world networks, show both accurate. From experiments, also demonstrated tends perform better networks heterogeneous distributions.