Research Center of Language Technology, Harbin Institute of Technology, Harbin, China.
Department of Mathematics, Harbin Institute of Technology, Harbin, China.
Artif Intell Med. 2019 Jan;93:43-49. doi: 10.1016/j.artmed.2018.05.001. Epub 2018 May 18.
Deep learning research on relation classification has achieved solid performance in the general domain. This study proposes a convolutional neural network (CNN) architecture with a multi-pooling operation for medical relation classification on clinical records and explores a loss function with a category-level constraint matrix. Experiments using the 2010 i2b2/VA relation corpus demonstrate these models, which do not depend on any external features, outperform previous single-model methods and our best model is competitive with the existing ensemble-based method.
深度学习在关系分类方面的研究已经在一般领域取得了坚实的成果。本研究提出了一种用于临床记录中医疗关系分类的卷积神经网络 (CNN) 架构，该架构采用了多池化操作，并探索了一种具有类别级约束矩阵的损失函数。使用 2010 年 i2b2/VA 关系语料库进行的实验表明，这些模型不依赖于任何外部特征，优于以前的单一模型方法，我们的最佳模型与现有的基于集成的方法具有竞争力。
