ABSTRACT
Bi-LSTM-CRF (Bi-Directional Long Short-Term Memory Conditional Random Field) model have good performance in Chinese medical Electronic Medical Records (EMRS) Named Entity Recognition (NER), However, Bi-LSTM-CRF model cannot make full use of the parallelism of GPU (Graphics Processing Unit) in massive medical records, and the neglect of word order features and semantic information in IDCNN(Iterated Dilated Convolutional Neural Networks) model leads to poor NER effect. Therefore, this paper proposes a BERT-IDCNN-CRF model. In this model, the two-way transformer pre training model BERT is used to fine tune the model parameters in the manual annotated corpus conforming to the BIOES (Begin Inside Outside End Single) standard. The text is learned in an unsupervised manner, and the semantic information of words is represented by word vectors, which can well represent the context semantics in the sentences of EMRS; The state characteristics of character sequences are learned through BERT model, and the sequence state scores obtained are input to the CRF layer. The CRF layer makes constraint optimization on the sequence state transition, and IDCNN has better recognition effect on convolutional coding of local entities. Experimental test results: the average accuracy, recall and F1 value of the BERT-IDCNN-CRF model are 94.5%, 93.8% and 94.1% respectively, which are increased by 4.8%, 4.3% and 3.6% respectively compared with the baseline model Word2Vec-BiLSTM-CRF. The experiment proves that the BERT-IDCNN-CRF model can better identify medical entities in electronic medical records.
- Jiang Xiang, Ma Jianxia, Yuan Hui. Named entity recognition in the field of ecological governance technology based on bilstm idcnn CRF model. computer applications and software, 2021, 38 (03): 134-141Google Scholar
- Collobert R, Weston J, Bottou L, Natural Language Processing (almost) from Scratch. Journal of Machine Learning Research, 2011, 12(1):2493-2537.Google ScholarDigital Library
- Strubell E , Verga P , Belanger D , Fast and Accurate Entity Recognition with Iterated Dilated Convolutions[C]// Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. 2017.Google Scholar
- Qiu J, Wang Q, Zhou Y, Fast and Accurate Recognition of Chinese Clinical Named Entities with Residual Dilated Convolutions// 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). IEEE Computer Society, 2018.Google Scholar
- Huang Z, Wei X, Kai Y. Bidirectional LSTM-CRF Models for Sequence Tagging. Computer Science, 2015.Google Scholar
- Research on Named Entity Recognition Based on Bi LSTM and CRF - Taking entities related to ecological governance technology as an example. University of Chinese Academy of Sciences, 2017Google Scholar
- Ma X, Hovy E. End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF.016.Google Scholar
- He Tao, Chen Jian, Wen yingyou. Research on entity recognition of electronic medical records based on bert-crf model. Computer and digital engineering, 2022, 50 (03): 639-643Google Scholar
- Liang Wentong, Zhu Yanhui, Zhan Fei, Ji Xiangbing. Named entity recognition of medical electronic medical records based on Bert. Journal of Hunan University of technology, 2020, 34 (04): 54-62Google Scholar
- Li Ni, Guan Huanmei, Yang Piao, Dong Wenyong. Chinese named entity recognition method based on bert-idcnn-crf. Journal of Shandong University (SCIENCE EDITION), 2020, 55 (01): 102-109Google Scholar
- Luo Xi, Xia Xianyun, An Ying, Chinese clinical entity recognition combined with multi head self attention mechanism and bilstm-crf. JOURNAL OF HUNAN UNIVERSITY (NATURAL SCIENCE EDITION), 2021, 48 (4): 45-55Google Scholar
- Chen Jian, he Tao, Wen yingyou, Ma Lintao. Entity recognition method of judicial documents based on Bert model. Journal of Northeastern University (NATURAL SCIENCE EDITION), 2020, 41 (10): 1382-1387Google Scholar
Index Terms
- Research on Application of Named Entity Recognition of Electronic Medical Records Based on BERT-IDCNN-CRF Model
Recommendations
A Hybrid Model Based on CRFs for Chinese Named Entity Recognition
ALPIT '08: Proceedings of the 2008 International Conference on Advanced Language Processing and Web Information TechnologyThis paper presents a hybrid model and the corresponding algorithm combining Conditional Random Fields (CRFs) with statistical methods to improve the performance of CRFs for the task of Chinese Named Entity Recognition (NER). CRFs has a good performance ...
Chinese mineral named entity recognition based on BERT model
AbstractMineral named entity recognition (MNER) is the extraction for the specific types of entities from unstructured Chinese mineral text, which is a prerequisite for building a mineral knowledge graph. MNER can also provide important data ...
Highlights- Present a BERT-based model for Chinese mineral named entity recognition.
- ...
Research on Named Entity Recognition of Traditional Chinese Medicine Electronic Medical Records
Health Information ScienceAbstractThe electronic medical record (EMR) is a patient’s individual medical record written by health care providers to describe the medical activities of patients. Named entity recognition (NER) of EMR is helpful to extract important information from a ...
Comments