iagoalves/teste_toxicidade_acl_250331

该数据集包含文本(text)和与其相关的多个标签，如目标群体(target_group)、是否为事实(factual?)、群体影响(ingroup_effect)、是否粗俗(lewd)、框架(framing)、预测群体(predicted_group)、刻板印象(stereotyping)、意图(intent)、AI评估的有害性(toxicity_ai)、人工评估的有害性(toxicity_human)、预测作者(predicted_author)、实际方法(actual_method)等。数据集分为训练集(train)，包含一定数量的示例。这些特征表明该数据集可能是用于文本分类、毒性检测或作者意图预测等NLP任务。

The dataset includes text and multiple associated labels such as target group, factual or not, ingroup effect, lewdness, framing, predicted group, stereotyping, intent, AI-assessed toxicity, human-assessed toxicity, predicted author, actual method, etc. The dataset is split into a training set (train) with a certain number of examples. These features suggest that the dataset might be used for NLP tasks such as text classification, toxicity detection, or author intent prediction.