Recurrent Neural Networks for Classifying Relations in Clinical Notes.
Journal of Biomedical Informatics(2017)
摘要
•RNN-LSTM for classifying three different types of relations from clinical notes.•Two different representations (sentence vs segment) for encoding the relations.•Comparable LSTM performance to state-of-the-art, with minimal feature engineering.•Word embeddings from clinical text more useful to the task than from general text.
更多查看译文
关键词
Natural language processing,Medical relation classification,Recurrent neural network,Long Short-Term Memory,Machine learning
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络