mteb/NusaParagraphEmotionClassification

NusaParagraphEmotionClassification是一个多语言情感分类数据集，包含10种印度尼西亚语言。这个数据集是基于NusaParagraph数据集构建的，用于情感分类任务，包括非虚构、虚构和书面文本。数据集包含了多种语言的训练、验证和测试数据，每个部分都有一定数量的样本和字符数。此外，还提供了数据集的描述性统计数据，包括样本数量、字符数量、文本长度、唯一文本数量、唯一标签数量以及每个标签的数量。

NusaParagraphEmotionClassification is a multi-class emotion classification dataset containing 10 Indonesian languages. It is built on the NusaParagraph dataset and is used for emotion classification tasks, including non-fiction, fiction, and written texts. The dataset includes training, validation, and test data for various languages, each with a certain number of samples and character count. Additionally, descriptive statistics of the dataset are provided, including the number of samples, number of characters, text length, number of unique texts, number of unique labels, and the count for each label.