.png){kind=icon}
.png){kind=icon}
Return
LogNER: Enhancing log semantics with LLM-driven entity recognition
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jss.2026.112892.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Establishes log entity recognition as a missing semantic dimension in existing log analysis. • Proposes the LogNER framework combining template-assisted and LLM-driven recognition. • Experiments on four heterogeneous datasets demonstrate strong effectiveness and generalization. • Ablation results confirm each core component’s essential contribution to overall performance.
Keywords:
Log entity recognition
Log analysis
Large Language Models
Entity recognition
LogNER framework
Journal
IF:
4.1
Papers:
5.4K
Citations:
8.4K