- 05-20李楠 Generalizing Biomedical Relation Classification with Neural Adversarial Domain Adaptation
- 04-23李青青 Deep learning for drug-drug interaction extraction
- 04-15刁宇峰 AAAI2018中的自注意力机制(Self-attention Mechanism)
- 04-15王治政 Label informed Attributed Network Embedding
- 03-24罗凌 自然语言处理中的自注意力机制（Self-attention Mechanism）
- 01-04任玉琪 Capsule间的动态路由
2017年9月27日，接到编辑部通知，博士生郑巍的论文“An attention-based effective neural model for drug-drug interactions extraction" 被BMC Bioinformatics录用，该期刊属于CCF推荐列表的C类期刊。
Here's part of the paper:
Drug-drug interactions (DDIs) often bring unexpected side effects. The clinical recognition of DDIs is a crucial issue for both patient safety and health care cost control. However, although text-mining-based DDIs systems explore various methods, classification performances of DDIs with long and complex sentences are still unsatisfactory. Methods In this study, we propose an effective model that classifies DDIs from literature by introducing an attention mechanism to the recurrent neural network with long shortterm memory (LSTM) units. In our approach, first, a candidate-drugs-oriented input attention acting on word embedding vectors automatically learns which words are more influential for a given drug pair. Next, the inputs merging the position and POS embedding vectors are passed to a bidirectional LSTM layer, whose outputs at the last time step represent the high-level semantic information of the whole sentence. Finally, a softmax layer performs DDIs classification. Results Experimental results on the DDIExtraction 2013 corpus show that our system achieves the best detecting and classification performance (84.0% and 77.3%, respectively) compared with other state-of-the-art methods. Especially for the Medline-2013 dataset with long and complex sentences, our F-score far outperforms top-ranking systems by 12.6%.
Our approach effectively improves performances of DDIs tasks. Experimental analysis demonstrates that our model has better performances for recognizing not only closerange but also long-range patterns among words, especially for long, complex and compound sentences.