Unveiling Influence in Networks: A Novel Centrality Metric and Comparative Analysis through Graph-Based Models

Area of Science:

• Network Science
• Computational Social Science
• Graph Theory

Background:

• Identifying influential nodes in social networks is crucial for managing information dissemination and disease transmission.
• Centrality measures are key tools for quantifying node influence in network analysis.
• Existing methods often focus on local or global network structures, with some combining both.

Purpose of the Study:

• To propose a novel centrality measure that integrates node degree with broader network path features.
• To evaluate the proposed measure's effectiveness against traditional centrality metrics.
• To validate the measure's utility in predicting real-world network dynamics, such as disease spread.

Main Methods:

• Developed a new centrality measure emphasizing node degree and incorporating network path information.
• Performed correlation analyses (Spearman, Pearson) using seven standard network datasets.
• Utilized the susceptible-infected-recovered (SIR) model to simulate virus spread and assess node influence capacity.

Main Results:

• The novel centrality measure demonstrated significant correlative efficacy compared to established metrics across diverse datasets.
• Validation using the SIR model confirmed the measure's ability to identify nodes with high infection potential.
• The proposed method shows robust performance in identifying influential actors in various network structures.

Conclusions:

• The new centrality measure offers an effective approach for identifying influential nodes in social networks.
• This method enhances the understanding of network structures and their impact on information and disease propagation.
• The findings have implications for targeted interventions in network-based scenarios.