Joint-NLU
Joint-NLU copied to clipboard
Published
20 hours ago •
InsaneLife
InsaneLife
Joint Slot Filling and Intent Prediction Use Attention and Slot Gate. NER, Intent classification
use slot-gated model for joint Slot Filling and Intent Prediction. and add CRF layer.
slot-gete论文解读
https://blog.csdn.net/shine19930820/article/details/83052300
CRF
crf开关：--use_crf
预训练
预训练词向量加入之后，效果反而变差了，Semantic Acc: 27.40左右。具体原因不明，有了解过的请告知，感谢。
后期加入bert实验。
实验结果
| ATIS数据集 | crf | softmax |
|---|---|---|
| Slot f1 | val: 97.39 ; test: 95.15 | val: 97.0; test:95.08 |
| Intent Acc | val: 98.2; test: 95.41 | val: 97.6; test: 95.52 |
| Semantic Acc | val: 90.0; test: 83.42 | val: 89.0 ; test: 83.53 |
Environment
tensorflow 1.11 python 3.5
the origin code : https://github.com/MiuLab/SlotGated-SLU
Reference
- Slot-Gated Modeling for Joint Slot Filling and Intent Prediction : http://www.aclweb.org/anthology/N18-2118
Metadata
37
Stars
13
Forks
Watchers
Owner
InsaneLife
Metadata
Joint Slot Filling and Intent Prediction Use Attention and Slot Gate. NER, Intent classification