Named Entity Recognition Model Based on the Fusion of Word Vectors and Category Vectors

Named entity recognition (NER) in deep learning mode heavily relies on the processing and analysis of text vectors. This paper introduces an NER model based on deep learning techniques, emphasizing the fusion of word vectors and category vectors to enhance text processing and analysis capabilities....

Full description

Saved in:
Bibliographic Details
Main Authors: Yang Zhou, Haoyang Zeng, Wei Zhang, Yuguang Zhang
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10806666/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850071634963070976
author Yang Zhou
Haoyang Zeng
Wei Zhang
Yuguang Zhang
author_facet Yang Zhou
Haoyang Zeng
Wei Zhang
Yuguang Zhang
author_sort Yang Zhou
collection DOAJ
description Named entity recognition (NER) in deep learning mode heavily relies on the processing and analysis of text vectors. This paper introduces an NER model based on deep learning techniques, emphasizing the fusion of word vectors and category vectors to enhance text processing and analysis capabilities. The model consists of two core components: a Fusion Vectors Generation (FVG) module and an NER module. In the FVG module, word vectors are derived from the BERT pre-training model, encapsulating the semantic information of the text. Additionally, category vectors are introduced to represent the entity type associated with the text. The relationship between word vectors and category vectors is learned using the BERT-BiLSTM architecture, and fusion vectors are then generated by combining these two types of vectors. On the other hand, the NER module utilizes the BiLSTM-CRF structure to extract contextual features from the fusion vectors and optimize the predicted label sequences. Consequently, a mapping between the fusion vectors and classification labels is established, facilitating the NER process. Validation on three NER datasets&#x2014;Weibo, Youku, and Chinese literature&#x2014;demonstrates that the proposed model is both feasible and effective. Compared to a baseline model that solely relies on word vectors, the proposed model exhibits enhanced NER performance, achieving improvements in <inline-formula> <tex-math notation="LaTeX">$F1$ </tex-math></inline-formula> scores by 5.05%, 1.53%, and 1.81% respectively on the Weibo, Youku, and Chinese literature datasets.
format Article
id doaj-art-4f0338c5a7564d0b885bbbae90530d37
institution DOAJ
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-4f0338c5a7564d0b885bbbae90530d372025-08-20T02:47:15ZengIEEEIEEE Access2169-35362024-01-011219465719466810.1109/ACCESS.2024.351930610806666Named Entity Recognition Model Based on the Fusion of Word Vectors and Category VectorsYang Zhou0https://orcid.org/0000-0002-8384-7842Haoyang Zeng1https://orcid.org/0009-0007-9771-3152Wei Zhang2https://orcid.org/0000-0003-0356-9561Yuguang Zhang3China Electronic Technology Cyber Security Co. Ltd., Chengdu, ChinaChina Electronic Technology Cyber Security Co. Ltd., Chengdu, ChinaChina Electronic Technology Cyber Security Co. Ltd., Chengdu, ChinaChina Electronic Technology Cyber Security Co. Ltd., Chengdu, ChinaNamed entity recognition (NER) in deep learning mode heavily relies on the processing and analysis of text vectors. This paper introduces an NER model based on deep learning techniques, emphasizing the fusion of word vectors and category vectors to enhance text processing and analysis capabilities. The model consists of two core components: a Fusion Vectors Generation (FVG) module and an NER module. In the FVG module, word vectors are derived from the BERT pre-training model, encapsulating the semantic information of the text. Additionally, category vectors are introduced to represent the entity type associated with the text. The relationship between word vectors and category vectors is learned using the BERT-BiLSTM architecture, and fusion vectors are then generated by combining these two types of vectors. On the other hand, the NER module utilizes the BiLSTM-CRF structure to extract contextual features from the fusion vectors and optimize the predicted label sequences. Consequently, a mapping between the fusion vectors and classification labels is established, facilitating the NER process. Validation on three NER datasets&#x2014;Weibo, Youku, and Chinese literature&#x2014;demonstrates that the proposed model is both feasible and effective. Compared to a baseline model that solely relies on word vectors, the proposed model exhibits enhanced NER performance, achieving improvements in <inline-formula> <tex-math notation="LaTeX">$F1$ </tex-math></inline-formula> scores by 5.05%, 1.53%, and 1.81% respectively on the Weibo, Youku, and Chinese literature datasets.https://ieeexplore.ieee.org/document/10806666/BERTnamed entity recognition (NER)bi-directional long short term memory networks (BiLSTM)category vectorsword vectors
spellingShingle Yang Zhou
Haoyang Zeng
Wei Zhang
Yuguang Zhang
Named Entity Recognition Model Based on the Fusion of Word Vectors and Category Vectors
IEEE Access
BERT
named entity recognition (NER)
bi-directional long short term memory networks (BiLSTM)
category vectors
word vectors
title Named Entity Recognition Model Based on the Fusion of Word Vectors and Category Vectors
title_full Named Entity Recognition Model Based on the Fusion of Word Vectors and Category Vectors
title_fullStr Named Entity Recognition Model Based on the Fusion of Word Vectors and Category Vectors
title_full_unstemmed Named Entity Recognition Model Based on the Fusion of Word Vectors and Category Vectors
title_short Named Entity Recognition Model Based on the Fusion of Word Vectors and Category Vectors
title_sort named entity recognition model based on the fusion of word vectors and category vectors
topic BERT
named entity recognition (NER)
bi-directional long short term memory networks (BiLSTM)
category vectors
word vectors
url https://ieeexplore.ieee.org/document/10806666/
work_keys_str_mv AT yangzhou namedentityrecognitionmodelbasedonthefusionofwordvectorsandcategoryvectors
AT haoyangzeng namedentityrecognitionmodelbasedonthefusionofwordvectorsandcategoryvectors
AT weizhang namedentityrecognitionmodelbasedonthefusionofwordvectorsandcategoryvectors
AT yuguangzhang namedentityrecognitionmodelbasedonthefusionofwordvectorsandcategoryvectors