JakubBilski/rap-vs-country-five-million-song-lyrics-subset

该数据集是一个音乐相关数据集，包含90,000个训练样本和10,000个测试样本，总大小约为186.68 MB。数据集特征包括标题、标签、艺术家、年份、浏览量、特征、歌词和ID，所有字段均以字符串格式存储。这些特征可能用于音乐信息检索、歌词分析或推荐系统等任务，但具体应用场景未在README中说明。数据集以标准分割形式提供，便于机器学习模型的训练和评估。This is a music-related dataset consisting of 90,000 training samples and 10,000 test samples, with an overall size of approximately 186.68 MB. The dataset contains the following fields: title, tags, artist, year, playback count, features, lyrics, and ID, all stored in string format. These fields can be utilized for tasks including music information retrieval, lyric analysis or recommendation systems, but the specific application scenarios are not specified in the README. The dataset is provided in a standard train-test split format to facilitate the training and evaluation of machine learning models.