Using Local Improved Structural Holes Method to Identify Key Nodes in Complex Networks

Abstract

In complex networks, it is significant to rank the nodes according to their importance. In this paper we present an algorithm based on an improved Structural Holes method to identify the key nodes of a complex network. Since our approach does not need to consider the global structure of a network but only consider the number of one node's neighbors and it's next nearest neighbors, the nodes importance can be calculated with local information of a complex network. Experimental results of ARPA net show that our method is better than some important ranking measures such as betweenness, degree or closeness. It is very useful for evaluating the key nodes in large scale and complicated networks, in which evaluation of nodes importance is almost impossible to calculate with global information.

Epidemic Model-based Network Influential Node Ranking Methods: A Ranking Rationality Perspective

“…A smaller network constraint index indicates a bigger structure hole and higher importance of the node. In the structural hole theory, the important nodes obtained by the method have a larger propagation range in the network, and show higher correctness than the comparison methods under the indicators of KCC and PC [43,45,128].…”

“…ERM [130], LISH [45], ISH [128], E-Burt [43], Bao et al [ Influence measures (PNI: The properties of the target node itself, INMN: The influence of the number of multi-hop neighbors, IMN: The influence among multi-hop neighbors, IAN: The influence of all nodes in the network)…”

“…However, classic structural hole method considers only the nearest neighbors. Local Improved Structural Holes (LISH) [45] and Improved Structural Holes (ISH) [128] take 2-hop neighbors into consideration. E-Burt [43] proposed the entropic degree concept and modified structural hole method.…”

“…Among them, the local method only applies the topological characteristics of nodes themselves in the network, so its limited use is to accurately identify the importance of nodes in large networks [17,19,22,31,47,83,84,86,92,96,114,119]. The semilocal method and the global method further consider the topological structure of the area near the node or even the entire network, so they can well identify important nodes in the network [3,16,24,34,43,45,52,103,128,130]. Based on the local method, the semi-local method and the global method synthesized to conform to the topological characteristics of a certain type of network and identify important nodes, the hybrid method would accumulate their index values as the importance of nodes [4,20,30,32,67,68,70,71,73,76,77,84,97,98,105,111,116,125,131,133,143,144,…”