[EMNLP2017]Context-Aware Representations for Knowledge Base Relation Extraction(short paper))
2017-10-29 17:03
656 查看
在上面给出的例子(1)中,任务是判断e1和e2的关系,在进行判断的时候,借助了e1和e3的关系. e1和e3的关系是在判断e1和e2之前已经识别出来.那么又引出了一个问题:如何表示e1和e3的关系?
图1给出的是实体e1和e2实体之间的关系判定,由下至上看此图,输入有两个序列(1)句子序列x1, x2, …, xn
(2)实体标签序列:标签B-∗, I-∗, L-∗, O, U-∗, ∗代表具体的实体类型,因为要判断e1和e2的关系,此时其他实体的标签全都打成O
在LSTM layer层的最后一个输出os就表示了实体e1和e2以及它们的关系
Figure2 就是利用其他关系判断两个目标实体之间的关系:
假设目前已经识别出的关系有两个r1和r2, 以及目前需要判定的目标实体e1和e2的关系rs, 三个关系的embedding表示o1, o2以及os就是由Figure 1的LSTM Layer层的最后一个输出得到.经由attention判断o1, o2哪个对os影响程度,利用公式:
最后os和oc串接后经由前馈神经网络得到关系:
相关文章推荐
- [EMNLP2017]Context-Aware Representations for Knowledge Base Relation Extraction
- Context-Aware Network Embedding for Relation Modeling
- [EMNLP2017]Deep Residual Learning for Weakly-Supervised Relation Extraction
- #Paper Reading# Online Knowledge-Based Model for Big Data Topic Extraction
- [AAAI2018]SEE:Syntax-aware Entity Embedding for Neural Relation Extraction
- CANE-Context-Aware Network Embedding for Relation Modeling论文学习
- [EMNLP2017]Adversarial Training for Relation Extraction
- Context-aware Ensemble of Multifaceted Factorization Models for Recommendation
- Reasoning With Neural Tensor Networks for Knowledge Base Completion
- CVPR2017: Learning Deep Context-aware Features over Body and Latent Parts for
- [EMNLP2017]Global Normalization of Convolutional Neural Networks for Joint Entity and Relation(阅读笔记)
- Context-Aware Pervasive Systems: Architectures for a New Breed of Applications
- 解决java.util.MissingResourceException: Can't find bundle for base name configContext, locale zh_CN
- 论文浅尝 | Distant Supervision for Relation Extraction
- 论文笔记:Distant Supervision for Relation Extraction via Piecewise Convolutional Neural Networks
- [Paper note] Joint Learning of Single-image and Cross-image Representations for Person Re-id.
- 【阅读笔记】Mining Concept Sequence from Large-Scale Search Logs for Context-Aware Query Suggestion
- 行为识别阅读笔记(paper + parted code):Beyond Frame-level CNN Saliency-Aware 3-D CNN with LSTM for Video Acti
- [EMNLP2015]Distant supervision for Relation Extraction via Piecewise Convolutional Neural Networks
- [EMNLP2017]Incorporating Relation Paths in Neural Relation Extraction