基于CTD-BLSTM的医疗领域中文命名实体识别模型  

Chinese Named Entity Recognition in Medical Field Using CTD-BLSTM Model

作　　者：祝锡永[1] 吴炀 刘崇 ZHU Xi-Yong;WU Yang;LIU Chong(School of Economics and Management,Zhejiang Sci-Tech University,Hangzhou 310018,China)

机构地区：[1]浙江理工大学经济管理学院,杭州310018

出　　处：《计算机系统应用》2020年第8期173-178,共6页Computer Systems & Applications

基　　金：国家自然科学基金(71501172);浙江省自然科学基金(LY15G010010)。

摘　　要：为在模型训练期间保留更多信息,用预训练词向量和微调词向量对双向长短期记忆网络(Bi-LSTM)神经模型进行扩展,并结合协同训练方法来应对医疗文本标注数据缺乏的情况,构建出改进模型CTD-BLSTM(Co-Training Double word embedding conditioned Bi-LSTM)用于医疗领域的中文命名实体识别.实验表明,与原始BLSTM与BLSTM-CRF相比,CTD-BLSTM模型在语料缺失的情况下具有更高的准确率和召回率,能够更好地支持医疗领域知识图谱的构建以及知识问答系统的开发.In order to retain more characteristic information in the training process,this study uses pre-training word vector and fine-tuning word vector to extend Bi-directional Long Short-Term Memory network(Bi-LSTM),and combines the co-training semi-supervision method to deal with the feature of sparse annotated text in the medical field.An improved model of Co-Training Double word embedding conditioned Bi-LSTM(CTD-BLSTM)is further proposed for Chinese named entity recognition.Experiments show that compared with the original BLSTM and BLSTM-CRF,the CTD-BLSTM model has higher accuracy and recall rate in the absence of corpora,the proposed method can better support the construction of medical knowledge graph and the development of knowledge answering system.

关 键 词：双向长短期记忆网络 协同训练 中文命名实体识别 问答系统 医疗领域 

分 类 号：R319[医药卫生—基础医学] TP391.1[自动化与计算机技术—计算机应用技术]