摘要
近年来,针对电子病历文本的研究受到越来越多的关注,而相关疾病预测模型很少注意到病历文本中记录独立分布的半结构化形式以及语义关系复杂的特点,故该文提出了一种基于加权层级注意力机制的辅助诊断方法,设计加权累加法将普通句向量转换为结构弱关联句向量,并构成词、句、文档层级结构注意力机制来提高模型结构学习能力,此外,设计监督层用于缓解语义关系复杂造成的学习偏置问题,以辅助模型的训练效果。在真实数据集中进行验证表明,该文模型优于当前主流的深度学习模型,取得了较好效果。
To capture the semi-structured information and the complex semantic relations in the medical record texts,this article proposes a disease prediction method based on a weighted hierarchical attention mechanism.The weighted accumulation method is designed to convert ordinary sentence vectors into structurally weakly related sentence vectors.A hierarchical structure attention mechanism is formed for the word,sentence,and document levels to improve the model.In addition,a supervision layer is constructed to alleviate the learning bias problem.Experiments on the real data set show the proposed model outperforms current deep learning models.
作者
单文琦
王波
黄青松
刘利军
黄冕
SHAN Wenqi;WANG Bo;HUANG Qingsong;LIU Lijun;HUANG Mian(School of Information Engineering and Automation,Kunming University of Science and Technology,Kunming,Yunnan 650500,China;Information Center of Yunnan Vocational College of Land and Resources,Kunming,Yunnan 652501,China;School of Information,Yunnan University,Kunming,Yunnan 650091,China;Key Laboratory of Computer Technology Application of Yunnan Province,Kunming University of Science and Technology,Kunming,Yunnan 650500,China)
出处
《中文信息学报》
CSCD
北大核心
2023年第1期97-103,共7页
Journal of Chinese Information Processing
基金
国家自然科学基金(81860318,81560296)。
关键词
累加法
注意力机制
层级结构
辅助诊断
accumulative method
attention mechanism
hierarchical structure
auxiliary diagnosis
