.png){kind=icon}
.png){kind=icon}
Return
HyDRA: Hyperbolic dual-geometry representation alignment for knowledge-aware recommendation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.neucom.2026.133848.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A dual-geometry recommendation framework is proposed to enhance robustness under noisy knowledge graphs. • Structural rationalization is introduced to suppress task-irrelevant relations and mitigate geometry-amplified noise. • Cross-geometry alignment coordinates Euclidean and hyperbolic representations for semantic consistency. • An adaptive fusion strategy balances geometric contributions for head and long-tail items.
Keywords:
Hyperbolic geometry
Dual-geometry framework
Knowledge-aware recommendation
Representation alignment
Noisy knowledge graphs
Journal
IF:
6.5
Papers:
2.5W
Citations:
6.5W