Abstract
Hierarchical text classification (HTC) is a challenging problem with two key issues: utilizing structural information and mitigating label imbalance. Recently, the unit-based approach generating unit-based feature representations has outperformed the global approach focusing on a global feature representation. Nevertheless, unit-based models using BCE and ZLPR losses still face static thresholding and label imbalance challenges. Those challenges become more critical in large-scale hierarchies. This paper introduces a novel hierarchy-aware loss function for unit-based HTC models: Hierarchy-aware Biased Bound Margin (HBM) loss. HBM integrates learnable bounds, biases, and a margin to address static thresholding and mitigate label imbalance adaptively. Experimental results on benchmark datasets demonstrate the superior performance of HBM compared to competitive HTC models.- Anthology ID:
- 2024.findings-acl.457
- Volume:
- Findings of the Association for Computational Linguistics: ACL 2024
- Month:
- August
- Year:
- 2024
- Address:
- Bangkok, Thailand
- Editors:
- Lun-Wei Ku, Andre Martins, Vivek Srikumar
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 7672–7682
- Language:
- URL:
- https://aclanthology.org/2024.findings-acl.457
- DOI:
- 10.18653/v1/2024.findings-acl.457
- Cite (ACL):
- Gibaeg Kim, SangHun Im, and Heung-Seon Oh. 2024. Hierarchy-aware Biased Bound Margin Loss Function for Hierarchical Text Classification. In Findings of the Association for Computational Linguistics: ACL 2024, pages 7672–7682, Bangkok, Thailand. Association for Computational Linguistics.
- Cite (Informal):
- Hierarchy-aware Biased Bound Margin Loss Function for Hierarchical Text Classification (Kim et al., Findings 2024)
- PDF:
- https://preview.aclanthology.org/dois-2013-emnlp/2024.findings-acl.457.pdf