Evolution mechanism of node importance based on the information about cascading failures in complex networks

Abstract

This paper mainly focuses on the evolution mechanism of node importance based on the information about cascading failures. Firstly, a novel node importance indicator is proposed according to the load turbulence of each node in the redistribution range based on a tunable load redistribution model. The indicator has two characteristics: one is that the failure consequence of the considered node can be clearly pointed out by its value, and the other is that the evolution mechanism of node importance can be analyzed with the factors of load redistribution rule, node capacity, and structural characteristics of the network. Then, an evaluation algorithm is presented. The indicator analytic formulas of Erdös-Rényi networks and Barabási-Albert networks are also presented respectively with the neighbor preferential and global preferential allocation rules. The experiments demonstrate the effectiveness and feasibility of the indicators and its algorithm, with which we also analyze the node importance evolution mechanism in-depth, namely how the not-so-great nodes in structure turns into the critical nodes to trigger cascading failure in complex networks.