A Lightweight Deep Learning-based Model for Ranking Influential Nodes in Complex Networks
Researchers have developed a new lightweight deep learning model called 1D-CGS for identifying influential nodes in complex networks. This hybrid model combines 1D convolutional neural networks with GraphSAGE to efficiently process topological features like node degree and neighbor degree. Experiments on various real-world networks show that 1D-CGS outperforms existing methods in ranking accuracy and runtime, demonstrating significant improvements in correlation and similarity metrics. AI
IMPACT Provides a more efficient method for identifying key entities in complex systems, potentially improving targeted interventions and network analysis.