Enhancing Text Classification by Graph Neural Networks With Multi-Granular Topic-Aware Graph
Text classification based on graph neural networks (GNNs) has been widely studied by virtue of its potential to capture complex and across-granularity relations among texts of different types from learning on a text graph. Existing methods typically construct text graphs based on words-documents to...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10054405/ |
| _version_ | 1811160938410147840 |
|---|---|
| author | Yongchun Gu Yi Wang Heng-Ru Zhang Jiao Wu Xingquan Gu |
| author_facet | Yongchun Gu Yi Wang Heng-Ru Zhang Jiao Wu Xingquan Gu |
| author_sort | Yongchun Gu |
| collection | DOAJ |
| description | Text classification based on graph neural networks (GNNs) has been widely studied by virtue of its potential to capture complex and across-granularity relations among texts of different types from learning on a text graph. Existing methods typically construct text graphs based on words-documents to capture relevant intra-class document representations among the same documents via words-words and words-documents propagation. However, a natural problem is that polysemy words in documents may become an information medium between documents of different categories, promoting heterophily information propagation. The performance of text classification will be somewhat constrained by this issue. This paper proposes a novel text classification method based on GNN from multi-granular topic-aware perspective, referred to as Text-MGNN. Specifically, topic nodes are introduced to build a triple node set of “word, document, topic,” and multi-granularity relations are modeled on a text graph for this triple node set. The introduction of topic nodes has three significant advantages. The first is to strengthen the propagation of topics, words, and documents. The second is to enhance class-aware representation learning. The final is to mitigate the effect of heterophily information caused by polysemy words. Extensive experiments are conducted on three real-world datasets. Results validate that our proposed method outperforms 11 baselines methods. |
| first_indexed | 2024-04-10T06:06:12Z |
| format | Article |
| id | doaj.art-74f558e642524125888d23a6b6303068 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-10T06:06:12Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-74f558e642524125888d23a6b63030682023-03-03T00:01:23ZengIEEEIEEE Access2169-35362023-01-0111201692018310.1109/ACCESS.2023.325010910054405Enhancing Text Classification by Graph Neural Networks With Multi-Granular Topic-Aware GraphYongchun Gu0https://orcid.org/0000-0002-7659-0081Yi Wang1Heng-Ru Zhang2https://orcid.org/0000-0001-9187-9847Jiao Wu3https://orcid.org/0000-0003-3181-0674Xingquan Gu4School of Mathematics, Sichuan University of Arts and Sciences, Dazhou, ChinaKey Laboratory of Intelligent Education Technology and Application of Zhejiang Province, Zhejiang Normal University, Jinhua, ChinaSchool of Computer Science, Southwest Petroleum University, Chengdu, ChinaCollege of Science, China Jiliang University, Hangzhou, ChinaCollege of Standardization, China Jiliang University, Hangzhou, ChinaText classification based on graph neural networks (GNNs) has been widely studied by virtue of its potential to capture complex and across-granularity relations among texts of different types from learning on a text graph. Existing methods typically construct text graphs based on words-documents to capture relevant intra-class document representations among the same documents via words-words and words-documents propagation. However, a natural problem is that polysemy words in documents may become an information medium between documents of different categories, promoting heterophily information propagation. The performance of text classification will be somewhat constrained by this issue. This paper proposes a novel text classification method based on GNN from multi-granular topic-aware perspective, referred to as Text-MGNN. Specifically, topic nodes are introduced to build a triple node set of “word, document, topic,” and multi-granularity relations are modeled on a text graph for this triple node set. The introduction of topic nodes has three significant advantages. The first is to strengthen the propagation of topics, words, and documents. The second is to enhance class-aware representation learning. The final is to mitigate the effect of heterophily information caused by polysemy words. Extensive experiments are conducted on three real-world datasets. Results validate that our proposed method outperforms 11 baselines methods.https://ieeexplore.ieee.org/document/10054405/Graph neural networkstext classificationtext graph construction |
| spellingShingle | Yongchun Gu Yi Wang Heng-Ru Zhang Jiao Wu Xingquan Gu Enhancing Text Classification by Graph Neural Networks With Multi-Granular Topic-Aware Graph IEEE Access Graph neural networks text classification text graph construction |
| title | Enhancing Text Classification by Graph Neural Networks With Multi-Granular Topic-Aware Graph |
| title_full | Enhancing Text Classification by Graph Neural Networks With Multi-Granular Topic-Aware Graph |
| title_fullStr | Enhancing Text Classification by Graph Neural Networks With Multi-Granular Topic-Aware Graph |
| title_full_unstemmed | Enhancing Text Classification by Graph Neural Networks With Multi-Granular Topic-Aware Graph |
| title_short | Enhancing Text Classification by Graph Neural Networks With Multi-Granular Topic-Aware Graph |
| title_sort | enhancing text classification by graph neural networks with multi granular topic aware graph |
| topic | Graph neural networks text classification text graph construction |
| url | https://ieeexplore.ieee.org/document/10054405/ |
| work_keys_str_mv | AT yongchungu enhancingtextclassificationbygraphneuralnetworkswithmultigranulartopicawaregraph AT yiwang enhancingtextclassificationbygraphneuralnetworkswithmultigranulartopicawaregraph AT hengruzhang enhancingtextclassificationbygraphneuralnetworkswithmultigranulartopicawaregraph AT jiaowu enhancingtextclassificationbygraphneuralnetworkswithmultigranulartopicawaregraph AT xingquangu enhancingtextclassificationbygraphneuralnetworkswithmultigranulartopicawaregraph |