Enhanced Hate Speech Annotations with Background Semantics - Data

This is the dataset associated to the paper: Enhancing Hate Speech Annotations with Background Semantics. ECAI 2024.The dataset is compiled in two CSV files: data and users. Data contains:Social media posts and background knowledge (of general hate speech terminology and specific terms to target individuals or groups).Hate speech and target group annotations in Phase 1, i.e. showing only the post.Hate speech and target group annotations in Phase 2, i.e. showing the post with highlighted terms and a side column with definitions.Users contains the demographic information of the annotators and experimental study conditions.Content WarningThis research aims to tackle hate speech and contains examples of triggering and harmful language. Risks associated with this study include feeling triggered or possibly harmed by viewing potentially toxic or hateful remarks and recalling unfavourable prior encounters. You may consider seeking support in the points: Stonewall (https://www.stonewall.org.uk/contact-stonewall’s-information-service), Q:alliance (https://www.qalliance.org.uk/).

本数据集与论文《通过背景语义增强仇恨言论标注》相关联，发表于 ECAI 2024。数据集由两个 CSV 文件组成：data 和 users。其中，data 文件包含：社交媒体帖子及其背景知识（包括一般仇恨言论术语以及针对个人或群体的特定术语）。第一阶段标注内容为仇恨言论及其目标群体，仅展示帖子内容。第二阶段标注内容为展示带有高亮术语的帖子及旁边的定义栏。users 文件包含标注者的人口统计信息及实验研究条件。内容警示：本研究旨在应对仇恨言论，并包含可能引起不适或有害语言的示例。参与本研究可能存在触发情绪或遭受伤害的风险，特别是当接触到潜在的毒素或仇恨言论以及回忆不愉快的过往经历时。如有需要，可寻求以下机构的支持：Stonewall（https://www.stonewall.org.uk/contact-stonewall’s-information-service）和 Q:alliance（https://www.qalliance.org.uk/）。