A Fast and Efficient Algorithm for Mining Top-k Nodes in Complex Networks.

Area of Science:

• Social network analysis
• Computational social science
• Network science

Background:

• Influence maximization is a key problem in social network analysis with theoretical and practical importance.
• Existing algorithms are typically propagation-based (high accuracy, slow) or topology-based (fast, unstable).

Purpose of the Study:

• To develop a novel topology-based algorithm for influence maximization that balances accuracy and speed.
• To improve upon the limitations of existing topology-based methods.

Main Methods:

• Proposed a novel topology-based algorithm utilizing a local index rank (LIR) metric.
• Evaluated the algorithm's performance on influence spread and running time against propagation-based methods.

Main Results:

• The LIR algorithm achieved influence spread comparable to, and sometimes exceeding, propagation-based algorithms.
• The LIR algorithm demonstrated a significant reduction in running time, being millions of times faster than propagation-based approaches.
• The algorithm showed stable performance in both Independent Cascade (IC) and Linear Threshold (LT) models.

Conclusions:

• The proposed local index rank (LIR) algorithm offers an efficient and effective solution for influence maximization in large networks.
• LIR provides a promising alternative to existing methods, offering a favorable trade-off between accuracy and computational cost.