MOROCO

该数据集是一个包含摩尔达维亚和罗马尼亚方言的新闻文本语料库，文本内容根据主题进行了标注。具体包含五个主题标签：文化、财经、政治、科学、体育以及科技。数据集的规模为21,719个训练样本，验证样本和测试样本各有6,000个。该数据集的任务是对新闻文本进行主题识别。

This dataset is a news text corpus covering Moldovan and Romanian dialects, with texts annotated by their respective topics. It includes six topic labels: culture, finance, politics, science, sports, and technology. The dataset consists of 21,719 training samples, plus 6,000 validation samples and 6,000 test samples respectively. The task of this dataset is topic identification for news texts.