返回
Temporal knowledge graphs forecasting based on explainable temporal relation tree-graph
DOI:10.1016/j.neunet.2026.108582.png)
摘要
En 中文
在真实世界的时序知识图谱中,实体之间的关系通常表现出复杂的时序动态。有效建模多跳时序关系链并实现可解释推理仍然是时序知识图谱预测中的核心挑战,我们提出的模型TRTL(基于时序关系树的时序知识图谱预测)解决了这些问题。为应对这些挑战,我们引入了一种新颖的推理框架,该框架基于两种互补的图结构:序列锚定图,用于捕捉时序区间和实体关系对齐;时序关系树图,用于将多跳关系链组织成可解释的树状推理路径。这些结构通过引入注意力机制的树型LSTM进行编码,使模型能够有效捕捉时序逻辑和长距离依赖。基于树的符号推理过程提供了可解释的证据,增强了预测的透明度和可靠性。在两个时序区间基准测试上的实验表明,TRTL显著优于现有的符号化模型。
期刊
IF:
6.3
论文数:
8.2K
被引数:
3.0W
机构
引用论文
CARM: Confidence-aware recommender model via review representation learning and historical rating behavior in the online platforms
NEUROCOMPUTING
IF6.5
Adaptive Modality Interaction Transformer for Multimodal Knowledge Graph Completion自适应模态交互Transformer用于多模态知识图谱补全
Aggregation or separation? Adaptive embedding message passing for knowledge graph completion聚合还是分离?用于知识图谱补全的自适应嵌入信息传递
PLEASING: Exploring the historical and potential events for temporal knowledge graph reasoning
NEURAL NETWORKS
IF6.3
Logical Rule-Based Knowledge Graph Reasoning: A Comprehensive Survey基于逻辑规则的知识图谱推理:一项全面综述
Mathematics
IF0
Multi-hop reasoning over paths in temporal knowledge graphs using reinforcement learning基于强化学习的时态知识图路径多跳推理
ESERNet: Learning spectrogram structure relationship for effective speech emotion recognition with swin transformer in classroom discourse analysis
NEUROCOMPUTING
IF6.5

