FounderShubham1729/medical-qa-datasets

该数据集集合包含多个与医疗和健康护理相关的问答任务数据集。数据集涵盖了临床知识、医学遗传学、专业医学等多个医学领域。其中，all-processed数据集是medical-meadow-*和chatdoctor_healthcaremagic数据集的合并；chatdoctor_healthcaremagic数据集中的Chat Doctor术语被替换为chatbot；medical_meadow_cord19数据集被抽样至50,000个样本；truthful-qa-*数据集用于评估模型在文本生成中的真实性，特别是在健康和营养相关问题上，共有55个健康问题和16个营养问题。这些数据集适用于医疗问答场景的模型训练和评估。

This dataset collection includes multiple datasets focused on medical and healthcare-related question-answering tasks. The datasets cover various medical domains such as clinical knowledge, medical genetics, professional medicine, and more. The all-processed dataset is a concatenation of medical-meadow-* and chatdoctor_healthcaremagic datasets; the term Chat Doctor in chatdoctor_healthcaremagic is replaced with chatbot; the medical_meadow_cord19 dataset is subsampled to 50,000 samples; the truthful-qa-* dataset is used to evaluate the truthfulness of models in text generation, particularly in health and nutrition-related questions, with 55 health questions and 16 nutrition questions. These datasets are suitable for training and evaluating models in medical question-answering scenarios.