Integrating local and global information to identify influential nodes in complex networks

• Main Authors: Mukhtar, Mohd. Fariduddin, Abas, Zuraida Abal, Samsu Baharuddin, Azhari, Norizan, Mohd. Natashah, Wan Fakhruddin, Wan Farah Wani, Minato, Wakisaka, Abdul Rasib, Amir Hamzah, Zainal Abidin, Zaheera, Abdul Rahman, Ahmad Fadzli Nizam, Hairol Anuar, Siti Haryanti
• Format: Article
• Language: English
• Published: Springer Nature 2023
• Subjects: H Social Sciences (General); Q Science (General)
• Online Access: http://eprints.utm.my/106849/1/WanFarahWani2023_IntegratingLocalandGlobalInformationtoIdentify.pdf; http://eprints.utm.my/106849/; http://dx.doi.org/10.1038/s41598-023-37570-7

Centrality analysis is a crucial tool for understanding the role of nodes in a network, but it is unclear how different centrality measures provide much unique information. To improve the identification of influential nodes in a network, we propose a new method called Hybrid-GSM (H-GSM) that combines the K-shell decomposition approach and Degree Centrality. H-GSM characterizes the impact of nodes more precisely than the Global Structure Model (GSM), which cannot distinguish the importance of each node. We evaluate the performance of H-GSM using the SIR model to simulate the propagation process of six real-world networks. Our method outperforms other approaches regarding computational complexity, node discrimination, and accuracy. Our findings demonstrate the proposed H-GSM as an effective method for identifying influential nodes in complex networks.