.png){kind=icon}
.png){kind=icon}
Return
Temporal knowledge graph reasoning with local-global evolutionary patterns
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2025.112643.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Simultaneously modeling local and global evolutionary patterns for temporal knowledge graph reasoning. • Dynamic subgraph sampling method samples relevant local evolutionary information when the elements of facts appear discontinuous. • The temporal fusion encoder integrates time embeddings to assign weights to different subgraphs modeling temporal localized evolutionary features. • Global historical information and a decay function to distinguish the different importance of historical entities and to model global evolutionary patterns.
Journal
IF:
7.6
Papers:
1.3W
Citations:
4.5W
Organization
No organization information available