A Multi-Task BERT-BiLSTM-AM-CRF Strategy for Chinese Named Entity Recognition

Tang, Xiaoyong; Huo, Yong; Xia, Meng; Long, Chengfeng

doi:10.1007/s11063-022-10933-3

Cited by 11 publications

(10 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Before experimenting, a dataset of online news domains is required, which is obtained by utilizing the ResumeNER corpus to identify the named entities of financial domain information in online news, thereby enabling sentiment monitoring based on it. Subsequently, training data proportions ranging from 20% to 50% are selected from the experimental corpus as training sets for comparison scheme models such as BiLSTM-CRF ( Luo et al, 2018 ), LM-LSTM-CRF ( Shi, 2022 ), Lite-LSTM ( Tang et al, 2022 ), and trigger matching networks (TMN) ( Lin et al, 2020 ), a concept and trigger matching network framework for entity triggers. The model proposed in this article is trained using 5% to 20% of the training data.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…In another study ( Kim et al, 2020 ), the classical LSTM+CRF structure was employed for NER. In a survey of Chinese NER ( Tang et al, 2022 ), a lattice-LSTM structure was proposed to mine better the character features in Chinese. Various word segmentation results are introduced into the model, and word information is remotely transmitted to nodes to form a “grid structure” to improve efficiency.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Named entity recognition and emotional viewpoint monitoring in online news using artificial intelligence

2024

PeerJ Computer Science

View full text Add to dashboard Cite

Network news is an important way for netizens to get social information. Massive news information hinders netizens to get key information. Named entity recognition technology under artificial background can realize the classification of place, date and other information in text information. This article combines named entity recognition and deep learning technology. Specifically, the proposed method introduces an automatic annotation approach for Chinese entity triggers and a Named Entity Recognition (NER) model that can achieve high accuracy with a small number of training data sets. The method jointly trains sentence and trigger vectors through a trigger-matching network, utilizing the trigger vectors as attention queries for subsequent sequence annotation models. Furthermore, the proposed method employs entity labels to effectively recognize neologisms in web news, enabling the customization of the set of sensitive words and the number of words within the set to be detected, as well as extending the web news word sentiment lexicon for sentiment observation. Experimental results demonstrate that the proposed model outperforms the traditional BiLSTM-CRF model, achieving superior performance with only a 20% proportional training data set compared to the 40% proportional training data set required by the conventional model. Moreover, the loss function curve shows that my model exhibits better accuracy and faster convergence speed than the compared model. Finally, my model achieves an average accuracy rate of 97.88% in sentiment viewpoint detection.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Named entity recognition and emotional viewpoint monitoring in online news using artificial intelligence

2024

PeerJ Computer Science

View full text Add to dashboard Cite

show abstract

“…Under the same experimental settings, we compared the performance of our EMLB model with the methods proposed by Nasar et al ( 2021 ), Fabregat et al ( 2023 ), Govindarajan et al ( 2023 ), Ke et al ( 2023 ), Laursen et al ( 2023 ), Tang et al ( 2023 ). After conducting multiple independent experiments, we used the average F1 score as the evaluation metric.…”

Section: Methodsmentioning

confidence: 99%

Integration of natural and deep artificial cognitive models in medical images: BERT-based NER and relation extraction for electronic medical records

Guo,

Liu,

Niu

2023

Front. Neurosci.

View full text Add to dashboard Cite

IntroductionMedical images and signals are important data sources in the medical field, and they contain key information such as patients' physiology, pathology, and genetics. However, due to the complexity and diversity of medical images and signals, resulting in difficulties in medical knowledge acquisition and decision support.MethodsIn order to solve this problem, this paper proposes an end-to-end framework based on BERT for NER and RE tasks in electronic medical records. Our framework first integrates NER and RE tasks into a unified model, adopting an end-to-end processing manner, which removes the limitation and error propagation of multiple independent steps in traditional methods. Second, by pre-training and fine-tuning the BERT model on large-scale electronic medical record data, we enable the model to obtain rich semantic representation capabilities that adapt to the needs of medical fields and tasks. Finally, through multi-task learning, we enable the model to make full use of the correlation and complementarity between NER and RE tasks, and improve the generalization ability and effect of the model on different data sets.Results and discussionWe conduct experimental evaluation on four electronic medical record datasets, and the model significantly out performs other methods on different datasets in the NER task. In the RE task, the EMLB model also achieved advantages on different data sets, especially in the multi-task learning mode, its performance has been significantly improved, and the ETE and MTL modules performed well in terms of comprehensive precision and recall. Our research provides an innovative solution for medical image and signal data.

show abstract

“…The evaluation corpus used in related research comes from English media organizations such as The Wall Street Times, The New York Times, and Wikipedia. The task of named entity recognition has stabilized and developed through a process of continuous improvement [4] . In the 1950s, researchers rst studied structured entities in papers and medical records.…”

Section: Introductionmentioning

confidence: 99%

Research on Chinese named entity recognition based on feature fusion

Lin,

Mao,

Zhang

2024

Preprint

View full text Add to dashboard Cite

Its core goal is to extract entities with specific meanings from text, and classify and label them. Recognizing entities in text can provide support for downstream tasks such as knowledge graphs, machine translation, and question answering systems, and therefore has important practical significance and research value. At present, the deep learning technology is used to classify Chinese names, and a satisfactory result is obtained. However, due to the characteristics and complexity of Chinese semantics, the current entity recognition methods of Chinese names still have some problems, such as the entity boundary is difficult to be divided, the structure is flexible, and the entity is easy to be confused. This project intends to start with the extraction of multi-source information of Chinese name entities, build semantic information based on multi-source semantic information, and build a semantic entity model based on feature fusion. Therefore, this paper proposes a method of Chinese name ontology recognition based on semantic analysis. Based on this, this paper conducts research on Chinese named entity recognition technology that uses multiple features for semantic enhancement. First, the research background and current situation of named entity recognition are analyzed and the various basic neural networks widely used in Chinese named entity recognition algorithms and the basic structure of mainstream models are introduced. Secondly, the entity boundaries faced by current Chinese named entity recognition are difficult to divide., flexible word formation, entity ambiguity and other difficulties, two Chinese named entity recognition models based on external semantic information enhancement are proposed to solve the above-mentioned problems existing in the Chinese named entity recognition task and further improve the three different models of MSRA, Resume and Weibo.

show abstract

A Multi-Task BERT-BiLSTM-AM-CRF Strategy for Chinese Named Entity Recognition

Cited by 11 publications

References 30 publications

Named entity recognition and emotional viewpoint monitoring in online news using artificial intelligence

Named entity recognition and emotional viewpoint monitoring in online news using artificial intelligence

Integration of natural and deep artificial cognitive models in medical images: BERT-based NER and relation extraction for electronic medical records

Research on Chinese named entity recognition based on feature fusion

Contact Info

Product

Resources

About