[NLU] Enhancing Dialogue Symptom Diagnosis with Global Attention and Symptom Graph
总述： 通过构造一种global attention的机制以及构造了symptom的图，提高了在医疗对话中，对于每一句话出现symptom的预测精度。以及提高了symptom inference（对于一种症状，病人是否有）的精度 贡献： 1.有一个开源的中文医疗对话数据集，大约两千多个对话，标注了BIO，做了symptom normalization。有四种病，大约162中症状 2.结合了一个gl
相关文章
- 1. [NLU] Improving Multi-turn Dialogue Modelling with Utterance ReWriter
- 2. 论文总结 | Dialogue Relation Extraction with Document-level Heterogeneous Graph Attention Networks
- 3. 论文笔记——Permutohedral-GCN: Graph Convolutional Networks with Global Attention
- 4. Survey of modern Fault Diagnosis methods in networks
- 5. Contextualized Graph Attention Network for Recommendation with Item Knowledge Graph
- 6. Paper Notes: Graph Attention Networks
- 7. 【论文解读 EMNLP 2019 | MOGANED】ED with Multi-Order Graph Convolution and Aggregated Attention
- 8. 【ICJAI2018】Commonsense Knowledge Aware Conversation Generation with Graph Attention
- 9. ICML2019 Self-Attention Graph Pooling
- 10. TextSpotter with Explicit Alignment and Attention
- 更多相关文章...
- • XSL-FO table-and-caption 对象 - XSL-FO 教程
- • W3C RDF and OWL 活动 - W3C 教程
- • RxJava操作符（七）Conditional and Boolean
- • 算法总结-股票买卖
相关标签/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
欢迎关注本站公众号,获取更多信息
相关文章
- 1. [NLU] Improving Multi-turn Dialogue Modelling with Utterance ReWriter
- 2. 论文总结 | Dialogue Relation Extraction with Document-level Heterogeneous Graph Attention Networks
- 3. 论文笔记——Permutohedral-GCN: Graph Convolutional Networks with Global Attention
- 4. Survey of modern Fault Diagnosis methods in networks
- 5. Contextualized Graph Attention Network for Recommendation with Item Knowledge Graph
- 6. Paper Notes: Graph Attention Networks
- 7. 【论文解读 EMNLP 2019 | MOGANED】ED with Multi-Order Graph Convolution and Aggregated Attention
- 8. 【ICJAI2018】Commonsense Knowledge Aware Conversation Generation with Graph Attention
- 9. ICML2019 Self-Attention Graph Pooling
- 10. TextSpotter with Explicit Alignment and Attention