Hierarchical knowledge distillation framework for efficient node influence prediction in large-scale complex networks

Abstract Node influence prediction is fundamental to epidemic control, viral marketing, and infrastructure resilience, yet traditional susceptible-infected-recovered (SIR) simulations require $$O(n \cdot R)$$ computational operations, rendering real-time applications infeasible for large-scale netwo...

Ausführliche Beschreibung

Bibliografische Detailangaben

Hauptverfasser:	Xiaomo Yu, Jiajia Liu, Ling Tang, Jie Mi, Long Long
Format:	Artigo
Sprache:	Inglês
Veröffentlicht:	Nature Portfolio 2026-05-01
Schriftenreihe:	Scientific Reports
Schlagworte:	Node influence prediction Knowledge distillation Graph neural networks Hierarchical teacher-student architecture LightGCN Complex networks
Online-Zugang:	https://doi.org/10.1038/s41598-026-47807-w
Tags:	Tag hinzufügen Keine Tags, Fügen Sie das erste Tag hinzu!