Demographics Should Not Be the Reason of Toxicity: Mitigating Discrimination in Text Classifications with Instance Weighting

Guanhua Zhang, Bing Bai, Junqi Zhang, Kun Bai, Conghui Zhu, Tiejun Zhao


Abstract
With the recent proliferation of the use of text classifications, researchers have found that there are certain unintended biases in text classification datasets. For example, texts containing some demographic identity-terms (e.g., “gay”, “black”) are more likely to be abusive in existing abusive language detection datasets. As a result, models trained with these datasets may consider sentences like “She makes me happy to be gay” as abusive simply because of the word “gay.” In this paper, we formalize the unintended biases in text classification datasets as a kind of selection bias from the non-discrimination distribution to the discrimination distribution. Based on this formalization, we further propose a model-agnostic debiasing training framework by recovering the non-discrimination distribution using instance weighting, which does not require any extra resources or annotations apart from a pre-defined set of demographic identity-terms. Experiments demonstrate that our method can effectively alleviate the impacts of the unintended biases without significantly hurting models’ generalization ability.
Anthology ID:
2020.acl-main.380
Volume:
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2020
Address:
Online
Editors:
Dan Jurafsky, Joyce Chai, Natalie Schluter, Joel Tetreault
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4134–4145
Language:
URL:
https://aclanthology.org/2020.acl-main.380
DOI:
10.18653/v1/2020.acl-main.380
Bibkey:
Cite (ACL):
Guanhua Zhang, Bing Bai, Junqi Zhang, Kun Bai, Conghui Zhu, and Tiejun Zhao. 2020. Demographics Should Not Be the Reason of Toxicity: Mitigating Discrimination in Text Classifications with Instance Weighting. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4134–4145, Online. Association for Computational Linguistics.
Cite (Informal):
Demographics Should Not Be the Reason of Toxicity: Mitigating Discrimination in Text Classifications with Instance Weighting (Zhang et al., ACL 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.acl-main.380.pdf
Video:
 http://slideslive.com/38928723
Code
 ghzhang233/Non-Discrimination-Learning-for-Text-Classification