FatimahEmadEldin/MycoBase-Large-Scale-Text-to-SQL

MycoBase是一个合成但生物学上准确的数据集，专为压力测试Text-to-SQL系统而设计。它代表了一个用于真菌研究的信息系统，涵盖从分类学、基因组学到形态学和培养的各个方面。数据集具有高模式复杂性，包含2,016个表和超过9,000个外键关系，数据量达320,270行真实真菌学数据。它包含数万个真实真菌物种名称、著名真菌学家（作者）和科学出版物标题，并与Spider和BIRD基准工具兼容。数据集生成基于MycoBase核心框架，确保在57个真菌目中的生物学准确性，每个表和列都模拟了真实世界真菌学研究工作流程。MycoBase is a synthetic but biologically accurate dataset designed for stress-testing Text-to-SQL systems. It represents a research information system for the study of fungi, covering everything from taxonomy and genomics to morphology and cultivation. The dataset features high schema complexity with 2,016 tables and over 9,000 foreign key relationships, and includes 320,270 rows of realistic mycology data. It incorporates tens of thousands of actual fungal species names, famous mycologists (authors), and scientific publication titles, and is compatible with Spider and BIRD benchmark tools. The dataset was generated using the MycoBase core framework, ensuring biological literacy across 57 fungal orders, with every table and column modeled after real-world research workflows in mycology.