BioASQ 2020 Challenge Dataset

该数据集专为生物医学领域的语义索引和问答而设计，特别是为了应对BioASQ挑战。它包含了不同类型的问题（如是非题、事实性问题、列表题等），并基于多种评估指标进行评价，如平均精度均值（MAP）和F1分数。在训练集中，该数据集包含了23,423个独特的代码，而在评估集中则包含了911篇文章。这一大规模生物医学语义索引和问答任务的数据集，为相关领域的研究提供了丰富的资源。This dataset is specifically designed for semantic indexing and question answering in the biomedical domain, particularly for the BioASQ challenge. It includes diverse types of questions, such as yes/no questions, factual questions, list-based questions and more, and is evaluated using multiple metrics including Mean Average Precision (MAP) and F1-score. The training set contains 23,423 unique codes, while the evaluation set comprises 911 articles. This large-scale dataset for biomedical semantic indexing and question answering tasks offers abundant resources for research in related fields.