Research on Key Nodes Identification based on Clustering Coefficient in Topological Networks

Jianxi Wang

School of Computer Science, Pingdingshan University, Pingdingshan, 467000, China

Abstract: Traditional identification methods of key nodes in network mainly study the unauthorized and un-directed network. The identification effect of nodes in network with changing topological relations is unstable, and the identification method has low accuracy and poor robustness. Aiming at the above problems, the key nodes identification method based on clustering coefficient is studied. Based on the topology network model, the aggregation coefficients of nodes in the network are calculated. The aggregation coefficient is taken as the node characteristic attribute and the improved KNN algorithm is used to identify the key nodes. Compared with the traditional node recognition method, the experimental results show that the recognition accuracy of the proposed method can reach 97.15% on average. Moreover, the proposed method has good robustness in application and can be applied in practice.

Keywords: Agglomeration coefficient; Topological network; Key nodes; Node identification; Improved KNN