Balancing Methods for Multi-label Text Classification with Long-Tailed Class Distribution

Yi Huang, Buse Giledereli, Abdullatif Köksal, Arzucan Özgür, Elif Ozkirimli


Abstract
Multi-label text classification is a challenging task because it requires capturing label dependencies. It becomes even more challenging when class distribution is long-tailed. Resampling and re-weighting are common approaches used for addressing the class imbalance problem, however, they are not effective when there is label dependency besides class imbalance because they result in oversampling of common labels. Here, we introduce the application of balancing loss functions for multi-label text classification. We perform experiments on a general domain dataset with 90 labels (Reuters-21578) and a domain-specific dataset from PubMed with 18211 labels. We find that a distribution-balanced loss function, which inherently addresses both the class imbalance and label linkage problems, outperforms commonly used loss functions. Distribution balancing methods have been successfully used in the image recognition field. Here, we show their effectiveness in natural language processing. Source code is available at https://github.com/blessu/BalancedLossNLP.
Anthology ID:
2021.emnlp-main.643
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
8153–8161
Language:
URL:
https://aclanthology.org/2021.emnlp-main.643
DOI:
10.18653/v1/2021.emnlp-main.643
Bibkey:
Cite (ACL):
Yi Huang, Buse Giledereli, Abdullatif Köksal, Arzucan Özgür, and Elif Ozkirimli. 2021. Balancing Methods for Multi-label Text Classification with Long-Tailed Class Distribution. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 8153–8161, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Balancing Methods for Multi-label Text Classification with Long-Tailed Class Distribution (Huang et al., EMNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.emnlp-main.643.pdf
Video:
 https://aclanthology.org/2021.emnlp-main.643.mp4
Code
 Roche/BalancedLossNLP +  additional community code
Data
Reuters-21578