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

Area of Science:

• Network Science
• Computational Social Science

Background:

• Centrality analysis is vital for network research but lacks methods to distinguish unique information from various measures.
• Existing models like Global Structure Model (GSM) struggle to differentiate node importance effectively.

Purpose of the Study:

• To develop an improved method for identifying influential nodes in complex networks.
• To enhance the precision of network analysis by combining established centrality techniques.

Main Methods:

• Proposed Hybrid-GSM (H-GSM) by integrating K-shell decomposition and Degree Centrality.
• Evaluated H-GSM performance using the SIR model on six real-world networks to simulate propagation processes.

Main Results:

• H-GSM demonstrated superior performance compared to other methods.
• Achieved better results in computational complexity, node discrimination, and accuracy.
• Effectively characterized node impact with greater precision than GSM.

Conclusions:

• Hybrid-GSM (H-GSM) is an effective and accurate method for identifying influential nodes.
• The integration of K-shell decomposition and Degree Centrality offers significant advantages in network analysis.
• H-GSM provides a more nuanced understanding of node influence in complex systems.