Named Entity Recognition Method Based on Multi-Feature Fusion
Nowadays, user-generated content has become a crucial channel for obtaining information and authentic feedback. However, due to the varying cultural and educational levels of online users, the content of online reviews often suffers from inconsistencies in specification and the inclusion of arbitrar...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/1/388 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841549382709673984 |
|---|---|
| author | Weidong Huang Xinhang Yu |
| author_facet | Weidong Huang Xinhang Yu |
| author_sort | Weidong Huang |
| collection | DOAJ |
| description | Nowadays, user-generated content has become a crucial channel for obtaining information and authentic feedback. However, due to the varying cultural and educational levels of online users, the content of online reviews often suffers from inconsistencies in specification and the inclusion of arbitrary information. Consequently, the task of extracting key information from online reviews has become a prominent area of research. This paper proposes a combined entity recognition model for online reviews, aiming to improve the accuracy of Named Entity Recognition (NER). Initially, the Non-negative Matrix Factorization (NMF) model is employed to perform thematic clustering on the review texts, and entity types are extracted based on the clustering results. Subsequently, we introduce an entity recognition model utilizing the pre-trained BERT model as an embedding layer, with BiLSTM and DGCNN incorporating residual connection and gating mechanisms as feature extraction layers. The model also leverages multi-head attention for feature fusion, and the final results are decoded using a Conditional Random Field (CRF) layer. The model achieves an F1 score of 86.8383% on a collected dataset of online reviews containing eight entity categories. Experimental results demonstrate that the proposed model outperforms other mainstream NER models, effectively identifying key entities in online reviews. |
| format | Article |
| id | doaj-art-d58ae83ce38f42d38ea2cafe98842832 |
| institution | Kabale University |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-d58ae83ce38f42d38ea2cafe988428322025-01-10T13:15:23ZengMDPI AGApplied Sciences2076-34172025-01-0115138810.3390/app15010388Named Entity Recognition Method Based on Multi-Feature FusionWeidong Huang0Xinhang Yu1School of Management, Nanjing University of Posts and Telecommunications, Nanjing 210003, ChinaSchool of Management, Nanjing University of Posts and Telecommunications, Nanjing 210003, ChinaNowadays, user-generated content has become a crucial channel for obtaining information and authentic feedback. However, due to the varying cultural and educational levels of online users, the content of online reviews often suffers from inconsistencies in specification and the inclusion of arbitrary information. Consequently, the task of extracting key information from online reviews has become a prominent area of research. This paper proposes a combined entity recognition model for online reviews, aiming to improve the accuracy of Named Entity Recognition (NER). Initially, the Non-negative Matrix Factorization (NMF) model is employed to perform thematic clustering on the review texts, and entity types are extracted based on the clustering results. Subsequently, we introduce an entity recognition model utilizing the pre-trained BERT model as an embedding layer, with BiLSTM and DGCNN incorporating residual connection and gating mechanisms as feature extraction layers. The model also leverages multi-head attention for feature fusion, and the final results are decoded using a Conditional Random Field (CRF) layer. The model achieves an F1 score of 86.8383% on a collected dataset of online reviews containing eight entity categories. Experimental results demonstrate that the proposed model outperforms other mainstream NER models, effectively identifying key entities in online reviews.https://www.mdpi.com/2076-3417/15/1/388named entity recognitiontopic analysisfeature fusion |
| spellingShingle | Weidong Huang Xinhang Yu Named Entity Recognition Method Based on Multi-Feature Fusion Applied Sciences named entity recognition topic analysis feature fusion |
| title | Named Entity Recognition Method Based on Multi-Feature Fusion |
| title_full | Named Entity Recognition Method Based on Multi-Feature Fusion |
| title_fullStr | Named Entity Recognition Method Based on Multi-Feature Fusion |
| title_full_unstemmed | Named Entity Recognition Method Based on Multi-Feature Fusion |
| title_short | Named Entity Recognition Method Based on Multi-Feature Fusion |
| title_sort | named entity recognition method based on multi feature fusion |
| topic | named entity recognition topic analysis feature fusion |
| url | https://www.mdpi.com/2076-3417/15/1/388 |
| work_keys_str_mv | AT weidonghuang namedentityrecognitionmethodbasedonmultifeaturefusion AT xinhangyu namedentityrecognitionmethodbasedonmultifeaturefusion |