基于BERT的中药材治疗胃病的命名实体识别  

Named Entity Recognition of Traditional Chinese Medicine for Gastropathy Based on BERT

作　　者：熊磊 XIONG Lei(School of Mechanical and Electrical Engineering,Gansu Agricultural University,Lanzou 730070,China)

机构地区：[1]甘肃农业大学机电工程学院,甘肃兰州730070

出　　处：《软件导刊》2025年第1期57-64,共8页Software Guide

摘　　要：中药材用于胃病治疗的命名实体识别,是中药材开发领域文本信息挖掘的重要任务,也是构建知识图谱最重要的基础任务。为了更好地实现对中药材治疗胃病实体的提取,实验设计了5个命名实体识别模型进行实验比较,在预训练层、神经网络层,输出层都进行了不同设计,选择了更适合的BERT-BILSTM-CRF模型。首先,通过BERT生成特征提取层BILSTM的词向量;其次,利用BILSTM获取文本前后两个方向的特征得到相关特征向量;最后,利用CRF进行解码、标签预测,并讨论了模型各部分对实验的影响。实验表明,所提模型在自创数据集上的准确率、召回率、F1值分别为85.20%、85.47%、85.33%,相较于现有方法表现较好。The named-entity recognition of Chinese medicinal materials used for gastropathy treatment is one of the important tasks of text in‐formation mining in the field of Chinese medicinal materials development,and is one of the most important basic tasks of building a Knowl‐edge graph.In order to better realize the extraction of Chinese medicinal materials for the treatment of gastropathy entities,five Named-entity recognition models were designed for experimental comparison,and different designs were carried out in the input layer,neural network layer,and output layer.Finally,a more suitable BERT-BILSTM-CRF model was chosen.Firstly,BERT is used to generate word vectors for the neu‐ral network BILSTM.Then,BILSTM is used to obtain text features in the front and back directions of the text,obtaining relevant feature vec‐tors.Finally,CRF is used for decoding and label prediction.The experiment showed that the accuracy,recall,and F1 values of the model used in the experiment were 85.20%,85.47%,and 85.33%,respectively,on the dataset created by oneself.And among the relevant literature that has been searched,the relevant label item indicator performs the best.

关 键 词：中药材胃病治疗 命名实体识别 深度学习 BERT BILSTM-CRF 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]