A novel voting measure for identifying influential nodes in complex networks based on local structure

Area of Science:

• Network Science
• Graph Theory
• Data Mining

Background:

• Identifying influential nodes is crucial for understanding network structure and function.
• Existing algorithms like VoteRank lack accuracy due to insufficient consideration of network topology.
• There is a need for improved methods that incorporate node attributes and neighborhood information.

Purpose of the Study:

• To propose an enhanced algorithm, Edge Weighted VoteRank (EWV), for more accurate influential node identification.
• To leverage node attributes and neighborhood structure to improve upon existing methods.
• To address limitations of previous algorithms, such as poor accuracy and monotonicity.

Main Methods:

• Developed the Edge Weighted VoteRank (EWV) algorithm, inspired by human voting behavior.
• Incorporated edge weights to represent node attractiveness to their first-order neighborhood.
• Introduced node similarity into the voting process and reduced voting ability in the second-order neighborhood to prevent clustering.

Main Results:

• EWV demonstrated superior performance compared to seven other algorithms across 12 real-world networks.
• The algorithm showed enhanced node differentiation ability and ranked list accuracy.
• Empirical results validate the effectiveness of EWV in identifying influential nodes.

Conclusions:

• The Edge Weighted VoteRank (EWV) algorithm offers a significant improvement for identifying influential nodes in complex networks.
• EWV's approach of integrating edge weights and node similarity enhances accuracy and effectiveness.
• This method provides a more robust tool for network analysis and understanding.