深度学习基础上的中医实体抽取方法研究 |
投稿时间:2018-09-28 点此下载全文 |
引用本文:张艺品,关贝,吕荫润,等.深度学习基础上的中医实体抽取方法研究[J].医学信息学杂志,2019,40(2):58-63 |
摘要点击次数: |
全文下载次数: |
|
基金项目:科技部国家重点研发计划重点专项(项目编号:2017YFB1002300);大数据驱动的中医智能辅助诊断服务系统课题一“多模态异构中医药大数据高效获取与资源库建设”(项目编号:2017YFB1002301)和课题三“基于深度学习的中医多尺度认知方法和辩证论治分析模型”(项目编号:2017YFB1002303 |
|
中文摘要:介绍命名实体识别及模型应用研究情况,以中医典籍作为数据源,采用深度学习方法,进行中医疾病、方剂、中草药等实体抽取,设计BiLSTM-CRF序列标注模型,构建中医典籍实验语料进行实验,结果表明该模型算法具有高度准确性。 |
中文关键词:知识图谱 实体抽取 中医 深度学习 |
|
Study on the Entity Extraction Method of Traditional Chinese Medicine on the Basis of Deep Learning |
|
|
Abstract:The paper introduces the study on named entity recognition and model application, conducts extraction of entities such as Traditional Chinese Medicine(TCM) diseases, prescription, Chinese herbal medicine, etc., by adoption of deep learning method and taking TCM classics as data sources, designs the model for sequencing tagging-BiLSTM-CRF. It also conducts experiments by building corpus of experiments in TCM classics. The result shows that the aforesaid model algorithm is of high accuracy. |
keywords:knowledge graph entity extraction Traditional Chinese Medicine(TCM) deep learning |
查看全文 查看/发表评论 下载PDF阅读器 |