.png){kind=icon}
Return
Progressive alternating attribute-Structure optimization for multiplex heterogeneous graphs
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2026.131495.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Introduces an alternating framework for structure refinement and attribute completion. • Proposes dynamic structure learning guided by attribute similarity. • Designs a multi-hop attribute completion scheme with learnable hop-specific weights. • Employs a progressive alternating training strategy to stabilize co-optimization of both views. • Achieves state-of-the-art results on four real-world heterogeneous graph datasets.
Keywords:
attribute completion
structure refinement
heterogeneous graphs
multi-hop scheme
alternating optimization
Journal
IF:
7.5
Papers:
3.0W
Citations:
10.2W
Organization
Cited Papers
.png){kind=icon}
ShareSave
View more