TIB/TIBKAT-subject-indexing-dataset

TIBKAT学科索引数据集是一个多语言文本分类数据集，包含英语和德语两种语言。数据集规模在10万到100万之间。它被设计用于学科索引任务，并提供了训练集、开发集和两个测试集（盲测试集和黄金测试集），所有数据文件均采用Parquet格式存储。The TIBKAT Subject Indexing Dataset is a multilingual text classification dataset containing English and German. The dataset is designed for subject indexing tasks and includes training, development, and two test sets (blind test set and gold test set), all data files are stored in Parquet format.