OpenEA V1

该数据集从多语言的DBpedia中提取了实体对齐配对，涵盖了跨语言和单语言环境。具体来说，该数据集分为20%的训练集、10%的验证集以及70%的测试集，并包含了两种跨语言环境（英语到法语和英语到德语）以及两种单语言环境（DBpedia到Wikidata和DBpedia到YAGO）。此外，该数据集提供了两种规模，每种设置分别有15,000对和100,000对实体对齐数据，其任务目标是实体对齐。

This dataset extracts entity alignment pairs from multilingual DBpedia, covering both cross-lingual and monolingual scenarios. Specifically, it is split into 20% training set, 10% validation set, and 70% test set. It also includes two cross-lingual scenarios (English to French and English to German) and two monolingual scenarios (DBpedia to Wikidata and DBpedia to YAGO). Furthermore, this dataset offers two scales, with 15,000 and 100,000 entity alignment pairs respectively for each configuration, and its task objective is entity alignment.