A novel complex networks clustering algorithm based on the core influence of nodes.

Area of Science:

• Complex networks analysis
• Network topology
• Sociological network modeling

Background:

• Cluster structure is a key topological property in complex networks, arising from node heterogeneity.
• Existing network clustering algorithms often suffer from inaccuracy and slow convergence.
• Understanding and accurately identifying network clusters is crucial for complex network studies.

Purpose of the Study:

• To propose a novel and efficient network clustering algorithm.
• To improve the accuracy and speed of identifying cluster structures in complex networks.
• To simulate the natural process of cluster formation in sociological contexts.

Main Methods:

• The proposed algorithm calculates the core influence of nodes using betweenness centrality.
• It identifies core structures using discriminant functions, simulating sociological cluster formation.
• Remaining nodes are clustered using an optimization method for final structure determination.

Main Results:

• The algorithm demonstrates superior clustering accuracy compared to the Fast-Newman algorithm.
• It achieves faster convergence, improving computational efficiency.
• Experimental results confirm its effectiveness in precisely revealing real network cluster structures.

Conclusions:

• The core influence-based clustering algorithm offers enhanced accuracy and speed for complex networks.
• It provides a more precise method for uncovering the underlying cluster structures.
• This approach advances the study of complex networks by improving clustering methodologies.