K-HATERS

K-HATERS是一个针对韩语的仇恨言论检测数据集，由崇实大学人工智能融合学院创建，包含约192,158条新闻评论，每条评论都附有针对特定目标的冒犯性评分。该数据集是韩语中最大的攻击性语言数据集，首次提供三点评分量表上的目标特定评分，便于检测韩语中不同程度的仇恨表达。数据集的应用领域包括开发和评估仇恨言论检测模型，旨在解决在线环境中仇恨言论的识别和抑制问题。

K-HATERS is a Korean hate speech detection dataset created by the College of Artificial Intelligence Convergence, Soongsil University. It contains approximately 192,158 news comments, each annotated with a target-specific offensiveness score. As the largest aggressive language dataset for Korean, it is the first to offer target-specific offensiveness scores based on a three-point rating scale, facilitating the detection of hate speech of varying severity in Korean. The dataset is designed for developing and evaluating hate speech detection models, with the goal of addressing the identification and mitigation of hate speech in online environments.