wissamantoun/NanoNQArabic

NanoNQ阿拉伯语数据集是NanoBEIR多语言集合中NanoNQ基准的阿拉伯语版本。该数据集包含三个配置：文档集合(corpus)、查询语句(queries)和相关性判断(qrels)。文档集合包含了要搜索的文档，查询语句是用于搜索的查询，相关性判断是查询-文档对的相关性评分。这个数据集适用于信息检索模型的效率评估。

NanoNQ Arabic dataset is the Arabic version of the NanoNQ benchmark from the NanoBEIR multilingual collection. The dataset contains three configurations: document collection (corpus), search queries (queries), and relevance judgments (qrels). The document collection includes the documents to be searched through, the search queries are the queries used for searching, and the relevance judgments are relevance scores for query-document pairs. This dataset is suitable for efficient evaluation of information retrieval models.