ArPanEmo

ArPanEmo数据集是由塔伊夫大学信息科学与技术学院创建，专注于阿拉伯语在线内容中的细粒度情感识别，特别是在COVID-19疫情期间。该数据集包含11,128条来自Twitter、YouTube和在线报纸评论的帖子，均手工标注了10种情感类别或中性标签。数据集的创建过程涉及使用Python包收集与COVID-19相关的帖子，并通过半自动分类和手动标注进行情感分类。ArPanEmo数据集的应用领域包括开发机器学习和深度学习工具以识别文本中的情感，以及监控在线可疑行为或心理健康障碍。

The ArPanEmo dataset was developed by the College of Information Science and Technology at Taif University, focusing on fine-grained sentiment recognition in Arabic online content, particularly during the COVID-19 pandemic. This dataset comprises 11,128 posts sourced from Twitter, YouTube, and online newspaper comments, all manually annotated with 10 sentiment categories or a neutral label. The dataset creation process involved collecting COVID-19-related posts using Python packages, followed by sentiment classification via semi-automatic categorization and manual annotation. Application scenarios of the ArPanEmo dataset include developing machine learning and deep learning tools for text sentiment recognition, as well as monitoring online suspicious behaviors or mental health disorders.