Iterative Neighbour-Information Gathering for Ranking Nodes in Complex Networks

Area of Science:

• Network Science
• Computational Social Science
• Data Mining

Background:

• Node influence is crucial for understanding network dynamics, functions, and structures.
• A node's spreading ability is significantly influenced by its neighbors.

Purpose of the Study:

• To introduce an iterative neighbor information gathering (Ing) process for evaluating node influence in networks.
• To demonstrate the algorithm's adaptability to various network types and its relationship with existing centrality measures.

Main Methods:

• The Ing process iteratively combines neighbor information using a transformation matrix and a priori information.
• It assigns an Ing score to each node, evaluating its influence over multiple iterations.
• The algorithm's convergence properties and relationship to eigenvector centrality were analyzed.

Main Results:

• Simulations using the susceptible-infected-removed (SIR) model on real-world networks showed Ing provides more exact rankings compared to eight renowned centralities.
• The Ing process demonstrated effectiveness even without a priori information.
• An optimal iteration time was identified for maximizing the characterization of node influence.

Conclusions:

• The Ing process offers a robust and adaptable method for node influence ranking, bridging gaps among existing measures.
• The algorithm has potential applications in critical areas like infectious disease control and optimizing information dissemination strategies.