ViHOS: Hate Speech Spans Detection for Vietnamese
This addresses the challenge of moderating hate speech in Vietnamese social media, but it is incremental as it applies existing methods to a new dataset.
The authors tackled the problem of detecting hateful and offensive language spans in Vietnamese social media comments by creating the first human-annotated dataset (ViHOS) with 26k spans on 11k comments, and they found that XLM-R_Large achieved the best F1-scores in single and all spans detection, while PhoBERT_Large performed best in multiple spans detection.
The rise in hateful and offensive language directed at other users is one of the adverse side effects of the increased use of social networking platforms. This could make it difficult for human moderators to review tagged comments filtered by classification systems. To help address this issue, we present the ViHOS (Vietnamese Hate and Offensive Spans) dataset, the first human-annotated corpus containing 26k spans on 11k comments. We also provide definitions of hateful and offensive spans in Vietnamese comments as well as detailed annotation guidelines. Besides, we conduct experiments with various state-of-the-art models. Specifically, XLM-R$_{Large}$ achieved the best F1-scores in Single span detection and All spans detection, while PhoBERT$_{Large}$ obtained the highest in Multiple spans detection. Finally, our error analysis demonstrates the difficulties in detecting specific types of spans in our data for future research. Disclaimer: This paper contains real comments that could be considered profane, offensive, or abusive.