Inflating Topic Relevance with Ideology: A Case Study of Political Ideology Bias in Social Topic Detection Models

Meiqi Guo, Rebecca Hwa, Yu-Ru Lin, Wen-Ting Chung


Abstract
We investigate the impact of political ideology biases in training data. Through a set of comparison studies, we examine the propagation of biases in several widely-used NLP models and its effect on the overall retrieval accuracy. Our work highlights the susceptibility of large, complex models to propagating the biases from human-selected input, which may lead to a deterioration of retrieval accuracy, and the importance of controlling for these biases. Finally, as a way to mitigate the bias, we propose to learn a text representation that is invariant to political ideology while still judging topic relevance.
Anthology ID:
2020.coling-main.428
Volume:
Proceedings of the 28th International Conference on Computational Linguistics
Month:
December
Year:
2020
Address:
Barcelona, Spain (Online)
Venue:
COLING
SIG:
Publisher:
International Committee on Computational Linguistics
Note:
Pages:
4873–4885
Language:
URL:
https://aclanthology.org/2020.coling-main.428
DOI:
10.18653/v1/2020.coling-main.428
Bibkey:
Cite (ACL):
Meiqi Guo, Rebecca Hwa, Yu-Ru Lin, and Wen-Ting Chung. 2020. Inflating Topic Relevance with Ideology: A Case Study of Political Ideology Bias in Social Topic Detection Models. In Proceedings of the 28th International Conference on Computational Linguistics, pages 4873–4885, Barcelona, Spain (Online). International Committee on Computational Linguistics.
Cite (Informal):
Inflating Topic Relevance with Ideology: A Case Study of Political Ideology Bias in Social Topic Detection Models (Guo et al., COLING 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.coling-main.428.pdf
Code
 MeiqiGuo/COLING2020-BiasStudy