Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
1 file | 1.00 MB
Add to list
  1. Search results
  2. A multi-task framework with enhanced ...

A multi-task framework with enhanced hierarchical attention for sentiment analysis on Classical Chinese poetry : utilizing information from short lines

Quanqi Du (UGent) and Veronique Hoste (UGent)
(2024) Proceedings of the 4th International Conference on Natural Language Processing for Digital Humanities. p.113-122
Author
Organization
Project
Abstract
Classical Chinese poetry has a long history, dating back to the 11th century BC. By investigating the sentiment expressed in the poetry, we can gain more insights in the emotional life and history development in ancient Chinese culture. To help improve the sentiment analysis performance in the field of classical Chinese poetry, we propose to utilize the unique information from the individual short lines that compose the poem, and introduce a multi-task framework with hierarchical attention enhanced with short line sentiment labels. Specifically, the multi-task framework comprises sentiment analysis for both the overall poem and the short lines, while the hierarchical attention consists of word- and sentence-level attention, with the latter enhanced with additional information from short line sentiments. Our experimental results showcase that our approach leveraging more fine-grained information from short lines outperforms the state-of-the-art, achieving an accuracy score of 72.88% and an F1-macro score of 71.05%.
Keywords
Natural language processing, Sentiment analysis, Multi-tasking learning, Hierarchical attention, Classical Chinese Poetry

Downloads

  • Download
    A Multi-task Framework with Enhanced Hierarchical Attention for Sentiment Analysis on Classical Chinese Poetry Utilizing Information from Short Lines.pdf
    • full text (Published version)
    • |
    • open access
    • |
    • PDF
    • |
    • 1.00 MB

Citation

Please use this url to cite or link to this publication:

MLA
Du, Quanqi, and Veronique Hoste. “A Multi-Task Framework with Enhanced Hierarchical Attention for Sentiment Analysis on Classical Chinese Poetry : Utilizing Information from Short Lines.” Proceedings of the 4th International Conference on Natural Language Processing for Digital Humanities, edited by Mika Hämäläinen et al., Association for Computational Linguistics (ACL), 2024, pp. 113–22, doi:10.18653/v1/2024.nlp4dh-1.11.
APA
Du, Q., & Hoste, V. (2024). A multi-task framework with enhanced hierarchical attention for sentiment analysis on Classical Chinese poetry : utilizing information from short lines. In M. Hämäläinen, E. Öhman, S. Miyagawa, K. Alnajjar, & Y. Bizzoni (Eds.), Proceedings of the 4th International Conference on Natural Language Processing for Digital Humanities (pp. 113–122). https://doi.org/10.18653/v1/2024.nlp4dh-1.11
Chicago author-date
Du, Quanqi, and Veronique Hoste. 2024. “A Multi-Task Framework with Enhanced Hierarchical Attention for Sentiment Analysis on Classical Chinese Poetry : Utilizing Information from Short Lines.” In Proceedings of the 4th International Conference on Natural Language Processing for Digital Humanities, edited by Mika Hämäläinen, Emily Öhman, So Miyagawa, Khalid Alnajjar, and Yuri Bizzoni, 113–22. Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/2024.nlp4dh-1.11.
Chicago author-date (all authors)
Du, Quanqi, and Veronique Hoste. 2024. “A Multi-Task Framework with Enhanced Hierarchical Attention for Sentiment Analysis on Classical Chinese Poetry : Utilizing Information from Short Lines.” In Proceedings of the 4th International Conference on Natural Language Processing for Digital Humanities, ed by. Mika Hämäläinen, Emily Öhman, So Miyagawa, Khalid Alnajjar, and Yuri Bizzoni, 113–122. Association for Computational Linguistics (ACL). doi:10.18653/v1/2024.nlp4dh-1.11.
Vancouver
1.
Du Q, Hoste V. A multi-task framework with enhanced hierarchical attention for sentiment analysis on Classical Chinese poetry : utilizing information from short lines. In: Hämäläinen M, Öhman E, Miyagawa S, Alnajjar K, Bizzoni Y, editors. Proceedings of the 4th International Conference on Natural Language Processing for Digital Humanities. Association for Computational Linguistics (ACL); 2024. p. 113–22.
IEEE
[1]
Q. Du and V. Hoste, “A multi-task framework with enhanced hierarchical attention for sentiment analysis on Classical Chinese poetry : utilizing information from short lines,” in Proceedings of the 4th International Conference on Natural Language Processing for Digital Humanities, Miami, USA, 2024, pp. 113–122.
@inproceedings{01JD278M1JKB1E8XN22RAA7Z44,
  abstract     = {{Classical Chinese poetry has a long history, dating back to the 11th century BC. By investigating the sentiment expressed in the poetry, we can gain more insights in the emotional life and history development in ancient Chinese culture. To help improve the sentiment analysis performance in the field of classical Chinese poetry, we propose to utilize the unique information from the individual short lines that compose the poem, and introduce a multi-task framework with hierarchical attention enhanced with short line sentiment labels. Specifically, the multi-task framework comprises sentiment analysis for both the overall poem and the short lines, while the hierarchical attention consists of word- and sentence-level attention, with the latter enhanced with additional information from short line sentiments. Our experimental results showcase that our approach leveraging more fine-grained information from short lines outperforms the state-of-the-art, achieving an accuracy score of 72.88% and an F1-macro score of 71.05%.}},
  author       = {{Du, Quanqi and Hoste, Veronique}},
  booktitle    = {{Proceedings of the 4th International Conference on Natural Language Processing for Digital Humanities}},
  editor       = {{Hämäläinen, Mika and Öhman, Emily and Miyagawa, So and Alnajjar, Khalid and Bizzoni, Yuri}},
  isbn         = {{9798891761810}},
  keywords     = {{Natural language processing,Sentiment analysis,Multi-tasking learning,Hierarchical attention,Classical Chinese Poetry}},
  language     = {{eng}},
  location     = {{Miami, USA}},
  pages        = {{113--122}},
  publisher    = {{Association for Computational Linguistics (ACL)}},
  title        = {{A multi-task framework with enhanced hierarchical attention for sentiment analysis on Classical Chinese poetry : utilizing information from short lines}},
  url          = {{http://doi.org/10.18653/v1/2024.nlp4dh-1.11}},
  year         = {{2024}},
}
Altmetric
View in Altmetric