medalpaca/medical_meadow_medqa
收藏资源简介：
--- task_categories: - question-answering language: - en - zh tags: - medical --- # Dataset Card for MedQA ## Dataset Description - **Paper:** ### Dataset Summary This is the data and baseline source code for the paper: Jin, Di, et al. "What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams." From https://github.com/jind11/MedQA: >The data that contains both the QAs and textbooks can be downloaded from [this google drive folder](https://drive.google.com/file/d/1ImYUSLk9JbgHXOemfvyiDiirluZHPeQw/view?usp=sharing). A bit of details of data are explained as below: > > For QAs, we have three sources: US, Mainland of China, and Taiwan District, which are put in folders, respectively. All files for QAs are in jsonl file format, where each line is a data sample as a dict. The "XX_qbank.jsonl" files contain all data samples while we also provide an official random split into train, dev, and test sets. Those files in the "metamap" folders are extracted medical related phrases using the Metamap tool. > > For QAs, we also include the "4_options" version in for US and Mainland of China since we reported results for 4 options in the paper. > > For textbooks, we have two languages: English and simplified Chinese. For simplified Chinese, we provide two kinds of sentence spliting: one is split by sentences, and the other is split by paragraphs. ### Citation Information ``` @article{jin2020disease, title={What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams}, author={Jin, Di and Pan, Eileen and Oufattole, Nassim and Weng, Wei-Hung and Fang, Hanyi and Szolovits, Peter}, journal={arXiv preprint arXiv:2009.13081}, year={2020} } ```
任务类别： - 问答任务 语言： - 英语 - 汉语 标签： - 医学 # MedQA 数据集卡片 ## 数据集说明 - **论文：** ### 数据集概览 本数据集与基线源代码对应论文为：Jin Di等人发表的《该患者罹患何种疾病？源自医学考试的大规模开放域问答数据集》。 数据仓库地址：https://github.com/jind11/MedQA： > 包含问答数据与教科书文本的数据集可通过[此谷歌云端硬盘文件夹](https://drive.google.com/file/d/1ImYUSLk9JbgHXOemfvyiDiirluZHPeQw/view?usp=sharing)下载。数据集细节说明如下： > 针对问答数据，我们设有三个来源：美国、中国大陆以及中国台湾地区，分别置于对应文件夹中。所有问答数据文件均采用jsonl格式，每行均为一个以字典形式存储的样本。"XX_qbank.jsonl" 文件包含全部问答样本，同时我们还提供了官方随机划分的训练集、开发集与测试集。"metamap" 文件夹内的文件为使用Metamap工具提取的医学相关短语。 > 针对美国与中国大陆的问答数据，我们额外提供了「四选项」版本，与论文中报告的四选一实验设置保持一致。 > 针对教科书文本，我们提供了英文与简体中文两种语言版本。其中简体中文文本提供两种分句方式：一种按句子拆分，另一种按段落拆分。 ### 引用信息 @article{jin2020disease, title={What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams}, author={Jin, Di and Pan, Eileen and Oufattole, Nassim and Weng, Wei-Hung and Fang, Hanyi and Szolovits, Peter}, journal={arXiv preprint arXiv:2009.13081}, year={2020} }
MedQA数据集概述
数据集描述
- 任务类别： 问答（question-answering）
- 语言： 英语（en）、中文（zh）
- 标签： 医疗（medical）
数据集总结
- 来源论文： Jin, Di, et al. "What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams."
- 数据下载： 数据包含问答和教科书内容，可从此Google Drive文件夹下载。
数据详情
-
问答数据：
- 来源：美国、中国大陆、台湾地区，分别存放于不同文件夹。
- 格式：jsonl文件格式，每行代表一个数据样本。
- 版本：提供“4_options”版本，适用于美国和中国大陆。
-
教科书数据：
- 语言：英语和简体中文。
- 分割方式：简体中文提供按句子和按段落两种分割方式。
引用信息
@article{jin2020disease, title={What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams}, author={Jin, Di and Pan, Eileen and Oufattole, Nassim and Weng, Wei-Hung and Fang, Hanyi and Szolovits, Peter}, journal={arXiv preprint arXiv:2009.13081}, year={2020} }
