Hierarchical Sequence-to-Sequence Model for Multi-Label Text Classification
We propose a novel sequence-to-sequence model for multi-label text classification, based on a “parallel encoding, serial decoding” strategy. The model combines a convolutional neural network and self-attention in parallel as the encoder to extract fine-grained local neighborhoo...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8879472/ |
| _version_ | 1818855322150764544 |
|---|---|
| author | Zhenyu Yang Guojing Liu |
| author_facet | Zhenyu Yang Guojing Liu |
| author_sort | Zhenyu Yang |
| collection | DOAJ |
| description | We propose a novel sequence-to-sequence model for multi-label text classification, based on a “parallel encoding, serial decoding” strategy. The model combines a convolutional neural network and self-attention in parallel as the encoder to extract fine-grained local neighborhood information and global interaction information from the source text. We design a hierarchical decoder to decode and generate the label sequence. Our method not only gives full consideration to the interpretable fine-gained information in the source text but also effectively utilizes the information to generate the label sequence. We conducted a large number of comparative experiments on three datasets. The results show that the proposed model has significant advantages over the state-of-the-art baseline model. In addition, our analysis demonstrates that our model is competitive with the RNN-based Seq2Seq models and that it is more robust at handling datasets with a high label/sample ratio. |
| first_indexed | 2024-12-19T08:06:45Z |
| format | Article |
| id | doaj.art-4fa241d93401494fa01c76d2ef52c653 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-19T08:06:45Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-4fa241d93401494fa01c76d2ef52c6532022-12-21T20:29:45ZengIEEEIEEE Access2169-35362019-01-01715301215302010.1109/ACCESS.2019.29488558879472Hierarchical Sequence-to-Sequence Model for Multi-Label Text ClassificationZhenyu Yang0Guojing Liu1https://orcid.org/0000-0002-0649-2822School of Computer Science and Technology, Qilu University of Technology (ShanDong Academy of Sciences), Jinan, ChinaSchool of Computer Science and Technology, Qilu University of Technology (ShanDong Academy of Sciences), Jinan, ChinaWe propose a novel sequence-to-sequence model for multi-label text classification, based on a “parallel encoding, serial decoding” strategy. The model combines a convolutional neural network and self-attention in parallel as the encoder to extract fine-grained local neighborhood information and global interaction information from the source text. We design a hierarchical decoder to decode and generate the label sequence. Our method not only gives full consideration to the interpretable fine-gained information in the source text but also effectively utilizes the information to generate the label sequence. We conducted a large number of comparative experiments on three datasets. The results show that the proposed model has significant advantages over the state-of-the-art baseline model. In addition, our analysis demonstrates that our model is competitive with the RNN-based Seq2Seq models and that it is more robust at handling datasets with a high label/sample ratio.https://ieeexplore.ieee.org/document/8879472/Sequence-to-sequencemulti-label text classificationself-attentionhierarchical decoderattention mechanism |
| spellingShingle | Zhenyu Yang Guojing Liu Hierarchical Sequence-to-Sequence Model for Multi-Label Text Classification IEEE Access Sequence-to-sequence multi-label text classification self-attention hierarchical decoder attention mechanism |
| title | Hierarchical Sequence-to-Sequence Model for Multi-Label Text Classification |
| title_full | Hierarchical Sequence-to-Sequence Model for Multi-Label Text Classification |
| title_fullStr | Hierarchical Sequence-to-Sequence Model for Multi-Label Text Classification |
| title_full_unstemmed | Hierarchical Sequence-to-Sequence Model for Multi-Label Text Classification |
| title_short | Hierarchical Sequence-to-Sequence Model for Multi-Label Text Classification |
| title_sort | hierarchical sequence to sequence model for multi label text classification |
| topic | Sequence-to-sequence multi-label text classification self-attention hierarchical decoder attention mechanism |
| url | https://ieeexplore.ieee.org/document/8879472/ |
| work_keys_str_mv | AT zhenyuyang hierarchicalsequencetosequencemodelformultilabeltextclassification AT guojingliu hierarchicalsequencetosequencemodelformultilabeltextclassification |