Mitigating Bias in Conversations: A Hate Speech Classifier and Debiaser with Prompts
This work addresses bias in online discourse to promote inclusivity, but it appears incremental as it builds on existing methods for hate speech detection and debiasing.
The paper tackled bias in hate speech during conversations by proposing a two-step approach with a classifier and a debiasing component using prompts, resulting in a reduction in negativity on a benchmark dataset.
Discriminatory language and biases are often present in hate speech during conversations, which usually lead to negative impacts on targeted groups such as those based on race, gender, and religion. To tackle this issue, we propose an approach that involves a two-step process: first, detecting hate speech using a classifier, and then utilizing a debiasing component that generates less biased or unbiased alternatives through prompts. We evaluated our approach on a benchmark dataset and observed reduction in negativity due to hate speech comments. The proposed method contributes to the ongoing efforts to reduce biases in online discourse and promote a more inclusive and fair environment for communication.