SOS: Systematic Offensive Stereotyping Bias in Word Embeddings

Elsafoury, Fatma; Wilson, Steven R.; Katsigiannis, Stamos; Ramzan, Naeem

Abstract

Systematic Offensive stereotyping (SOS) in word embeddings could lead to associating marginalised groups with hate speech and profanity, which might lead to blocking and silencing those groups, especially on social media platforms. In this [id=stk]work, we introduce a quantitative measure of the SOS bias, [id=stk]validate it in the most commonly used word embeddings, and investigate if it explains the performance of different word embeddings on the task of hate speech detection. Results show that SOS bias exists in almost all examined word embeddings and that [id=stk]the proposed SOS bias metric correlates positively with the statistics of published surveys on online extremism. We also show that the [id=stk]proposed metric reveals distinct information [id=stk]compared to established social bias metrics. However, we do not find evidence that SOS bias explains the performance of hate speech detection models based on the different word embeddings.

Citation

Elsafoury, F., Wilson, S. R., Katsigiannis, S., & Ramzan, N. (2022, October). SOS: Systematic Offensive Stereotyping Bias in Word Embeddings. Presented at 29th International Conference on Computational Linguistics (COLING 2022), Gyeongju, Republic of Korea

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 29th International Conference on Computational Linguistics (COLING 2022)
• Start Date: Oct 12, 2022
• End Date: Oct 17, 2022
• Acceptance Date: Aug 16, 2022
• Publication Date: 2022-10
• Deposit Date: Aug 19, 2022
• Publicly Available Date: Apr 28, 2023
• Pages: 1263-1274
• Public URL: https://durham-repository.worktribe.com/output/1136394
• Publisher URL: https://aclanthology.org/2022.coling-1.108