Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A hybrid node ranking technique for finding influential nodes in complex social networks

A hybrid node ranking technique for finding influential nodes in complex social networks In today’s digital era, data pertaining to scientific research have attracted considerable attention of researchers. Data of scientific publications can be modeled in the form of networks such as citation networks, co-citation networks, collaboration networks, and others. Identification and ranking of important nodes in such networks is useful in many applications, such as finding most influential papers, most productive researchers, pattern of citation, and many more. The paper aims to discuss this issue.Design/methodology/approachA number of methods are available in literature for node ranking, and K-shell decomposition is one such method. This method categorizes nodes in different groups based on their topological position. The shell number of a node provides useful insights about the node’s importance in the network. It has been found that shells produced by the K-shell method need to be further refined to quantify the influence of the nodes aptly. In this work, a method has been developed, which ranks nodes by taking the core(s) as the origin and second-order neighborhood of a node as its immediate sphere of influence.FindingsIt is found that the performance of the proposed technique is either comparable or better than other methods in terms of correctness and accuracy. In case of assigning different ranks to nodes, the performance of the proposed technique is far more superior to existing methods. The proposed method can be used to rank authors, research articles, and fields of research.Originality/valueThe proposed method ranks nodes by their global position in a network as well as their local sphere of information. It leads to better quantification of a node’s impact. This method is found to be better in terms of accuracy and correctness. In case of assigning different ranks to nodes, the performance of the proposed technique is far more superior to existing methods. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Library Hi Tech Emerald Publishing

A hybrid node ranking technique for finding influential nodes in complex social networks

Library Hi Tech , Volume 40 (1): 17 – Feb 14, 2022

Loading next page...
 
/lp/emerald-publishing/a-hybrid-node-ranking-technique-for-finding-influential-nodes-in-twn0Pn3eX0
Publisher
Emerald Publishing
Copyright
© Emerald Publishing Limited
ISSN
0737-8831
DOI
10.1108/lht-01-2019-0019
Publisher site
See Article on Publisher Site

Abstract

In today’s digital era, data pertaining to scientific research have attracted considerable attention of researchers. Data of scientific publications can be modeled in the form of networks such as citation networks, co-citation networks, collaboration networks, and others. Identification and ranking of important nodes in such networks is useful in many applications, such as finding most influential papers, most productive researchers, pattern of citation, and many more. The paper aims to discuss this issue.Design/methodology/approachA number of methods are available in literature for node ranking, and K-shell decomposition is one such method. This method categorizes nodes in different groups based on their topological position. The shell number of a node provides useful insights about the node’s importance in the network. It has been found that shells produced by the K-shell method need to be further refined to quantify the influence of the nodes aptly. In this work, a method has been developed, which ranks nodes by taking the core(s) as the origin and second-order neighborhood of a node as its immediate sphere of influence.FindingsIt is found that the performance of the proposed technique is either comparable or better than other methods in terms of correctness and accuracy. In case of assigning different ranks to nodes, the performance of the proposed technique is far more superior to existing methods. The proposed method can be used to rank authors, research articles, and fields of research.Originality/valueThe proposed method ranks nodes by their global position in a network as well as their local sphere of information. It leads to better quantification of a node’s impact. This method is found to be better in terms of accuracy and correctness. In case of assigning different ranks to nodes, the performance of the proposed technique is far more superior to existing methods.

Journal

Library Hi TechEmerald Publishing

Published: Feb 14, 2022

Keywords: Technology infrastructure; Author influence; Complex networks; Hybrid ranking; Node ranking; Spreading capability

References