patrikgerard/measuring-hate-speech

该数据集主要用于情感分析和仇恨言论检测，涵盖了多个维度的标注信息，包括情感、尊重、侮辱、羞辱、地位、非人化、暴力、种族灭绝、攻击防御、仇恨言论等。此外，数据集还包含了标注者的个人信息，如性别、教育程度、收入、意识形态等。数据集分为训练集和测试集，分别包含108,444和27,112个样本。

This dataset is primarily used for sentiment analysis and hate speech detection, covering multiple dimensions of annotation information, including sentiment, respect, insult, humiliate, status, dehumanize, violence, genocide, attack_defend, hatespeech, etc. Additionally, the dataset includes annotators personal information such as gender, education level, income, ideology, etc. The dataset is divided into training and test sets, containing 108,444 and 27,112 samples respectively.