Anti-triangle centrality-based community detection in complex networks.

Area of Science:

• Network science
• Graph theory
• Data mining

Background:

• Community detection is crucial in analyzing complex networks (social, biological, technological).
• Existing algorithms often overlook the significant role of edges, focusing primarily on vertex properties.
• Understanding edge contributions is key to refining community detection methodologies.

Purpose of the Study:

• To introduce a novel community detection approach utilizing edge properties.
• To develop a method based on edge antitriangle centrality for uncovering community structures.
• To address the limitations of vertex-centric approaches in network analysis.

Main Methods:

• Introduction of edge centrality based on the antitriangle property.
• Development of the Edge Antitriangle Centrality with Isolated vertex handling (EACH) algorithm.
• Iterative removal of edges with the highest antitriangle centrality scores until scores reach zero.

Main Results:

• The EACH algorithm demonstrates superior accuracy in community detection compared to state-of-the-art methods.
• EACH exhibits lower computational complexity.
• The method successfully identifies inherent and consistent communities with a maximal diameter of four jumps.

Conclusions:

• Edge centrality, specifically the antitriangle property, offers a powerful new perspective for community detection.
• The EACH algorithm provides a parameter-free and effective tool for uncovering network communities.
• This approach enhances the understanding of network structures by incorporating edge-level analysis.