A text classification model for dynamic fusion of global and local features
Existing text classification models insufficiently utilize global and local information in texts, leading to subpar classification performance. In response to this issue, a text classification model called global and local features dynamic fusion (GLFDF) is proposed. The GLFDF model was initially de...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Journal of XPU
2024-08-01
|
| Series: | Xi'an Gongcheng Daxue xuebao |
| Subjects: | |
| Online Access: | http://journal.xpu.edu.cn/en/#/digest?ArticleID=1489 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849727341288226816 |
|---|---|
| author | ZHENG Wenjun ZHANG Shunxiang |
| author_facet | ZHENG Wenjun ZHANG Shunxiang |
| author_sort | ZHENG Wenjun |
| collection | DOAJ |
| description | Existing text classification models insufficiently utilize global and local information in texts, leading to subpar classification performance. In response to this issue, a text classification model called global and local features dynamic fusion (GLFDF) is proposed. The GLFDF model was initially designed with a dynamic fusion enhancement module to dynamically control the integration of global temporal features and local semantic features into specific positions of the word embedding matrix. Subsequently, the embedding matrix where global and local features fused was fed into a feature extraction module for further processing. Finally, the proposed model was tested on two public datasets, Ohsumed and THUCNews. Experimental results show that the GLFDF model achieves F1 scores of 63.24% and 92.50% on these datasets, respectively, surpassing other advanced text classification models and enhancing text classification performance. From the analysis of the ablation experiment, the dynamic fusion enhancement module fully makes the global temporal features and local semantic features of the text fused together, effectively solving the problem of insufficient use of global and local information in the text classification model. |
| format | Article |
| id | doaj-art-1cd4be0ebb59417c9cd20e8d3f1f9a98 |
| institution | DOAJ |
| issn | 1674-649X |
| language | zho |
| publishDate | 2024-08-01 |
| publisher | Editorial Office of Journal of XPU |
| record_format | Article |
| series | Xi'an Gongcheng Daxue xuebao |
| spelling | doaj-art-1cd4be0ebb59417c9cd20e8d3f1f9a982025-08-20T03:09:52ZzhoEditorial Office of Journal of XPUXi'an Gongcheng Daxue xuebao1674-649X2024-08-013849710510.13338/j.issn.1674-649x.2024.04.013A text classification model for dynamic fusion of global and local featuresZHENG Wenjun0ZHANG Shunxiang1School of Artificial Intelligence, Anhui University of Science and Technology, Huainan 232001, Anhui, ChinaSchool of Computer Science and Engineering, Anhui University of Science and Technology, Huainan 232001, Anhui, ChinaExisting text classification models insufficiently utilize global and local information in texts, leading to subpar classification performance. In response to this issue, a text classification model called global and local features dynamic fusion (GLFDF) is proposed. The GLFDF model was initially designed with a dynamic fusion enhancement module to dynamically control the integration of global temporal features and local semantic features into specific positions of the word embedding matrix. Subsequently, the embedding matrix where global and local features fused was fed into a feature extraction module for further processing. Finally, the proposed model was tested on two public datasets, Ohsumed and THUCNews. Experimental results show that the GLFDF model achieves F1 scores of 63.24% and 92.50% on these datasets, respectively, surpassing other advanced text classification models and enhancing text classification performance. From the analysis of the ablation experiment, the dynamic fusion enhancement module fully makes the global temporal features and local semantic features of the text fused together, effectively solving the problem of insufficient use of global and local information in the text classification model.http://journal.xpu.edu.cn/en/#/digest?ArticleID=1489natural language processingtext classificationdynamic fusiongating mechanismconvolutional neural networkrecurrent neural network |
| spellingShingle | ZHENG Wenjun ZHANG Shunxiang A text classification model for dynamic fusion of global and local features Xi'an Gongcheng Daxue xuebao natural language processing text classification dynamic fusion gating mechanism convolutional neural network recurrent neural network |
| title | A text classification model for dynamic fusion of global and local features |
| title_full | A text classification model for dynamic fusion of global and local features |
| title_fullStr | A text classification model for dynamic fusion of global and local features |
| title_full_unstemmed | A text classification model for dynamic fusion of global and local features |
| title_short | A text classification model for dynamic fusion of global and local features |
| title_sort | text classification model for dynamic fusion of global and local features |
| topic | natural language processing text classification dynamic fusion gating mechanism convolutional neural network recurrent neural network |
| url | http://journal.xpu.edu.cn/en/#/digest?ArticleID=1489 |
| work_keys_str_mv | AT zhengwenjun atextclassificationmodelfordynamicfusionofglobalandlocalfeatures AT zhangshunxiang atextclassificationmodelfordynamicfusionofglobalandlocalfeatures AT zhengwenjun textclassificationmodelfordynamicfusionofglobalandlocalfeatures AT zhangshunxiang textclassificationmodelfordynamicfusionofglobalandlocalfeatures |