.png){kind=icon}
返回
Gaussian embedding metric learning for few-shot knowledge graph completion
.png){kind=icon}
PRE
AI
DOI：10.1016/j.knosys.2026.115431.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
• We designed an entity neighbor selector to choose the most relevant neighbor by computing the correlation scores between neighbor relations and few-shot relations, reducing noise introduced by irrelevant neighbors during fusion. • We developed a Gaussian encoder to model the uncertainty of relation semantics using Gaussian distribution. We utilized a stochastic ManifoldE decoder to measure uncertainty and combined it with deterministic prototype scores to enhance measurement accuracy. • We conducted extensive experiments on the NELL and Wiki datasets, achieving overall better performance compared to baseline models. We also achieved optimal results across different few-shot scenarios, with predictions outperforming the best baselines on complex 1-N and N-N class relations.
Keyword:
few-shot learning
knowledge graph completion
Gaussian embedding
uncertainty modeling
relation semantics
期刊
K
IF:
7.6
论文数:
1.3W
被引数:
4.5W
机构
暂无机构信息
引用论文
Multi-view semantic enhancement model for few-shot knowledge graph completion面向少镜头知识图补全的多视图语义增强模型
PREAI
分享收藏分享收藏Relational multi-scale metric learning for few-shot knowledge graph completion基于关系多尺度度量学习的少样本知识图谱补全
PREAI
分享收藏Heterogeneous representation learning and matching for few-shot relation prediction.png)
PATTERN RECOGNITION
PATTERN RECOGNITIONIF7.6
PREAI
分享收藏Knowledge graph embedding by relational rotation and complex convolution for link prediction基于关系旋转和复卷积的知识图谱嵌入链接预测
PREAI
分享收藏分享收藏Complete feature learning and consistent relation modeling for few-shot knowledge graph completion用于少量知识图补全的完整特征学习和一致关系建模
PREAI
分享收藏A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects卷积神经网络综述: 分析、应用与展望
OAAI
OAAI分享收藏
查看更多内容