NL2SQL-BUGs
收藏资源简介:
NL2SQL-BUGs是由香港科技大学(广州)的研究团队创建的一个数据集,旨在帮助研究人员检测和分类NL2SQL转换中的语义错误。该数据集按照两级分类法对语义错误进行分类,涵盖了9个主要类别和多个子类别。每个实例都由自然语言查询、数据库模式以及相应的SQL查询组成,并对错误的SQL查询提供了详细的注释说明。该数据集可应用于自然语言处理和数据库管理领域,以促进NL2SQL技术的实用化和鲁棒性提升。
NL2SQL-BUGs is a dataset developed by the research team at The Hong Kong University of Science and Technology (Guangzhou), designed to assist researchers in detecting and classifying semantic errors occurring during NL2SQL conversions. The dataset adopts a two-level taxonomy for semantic error classification, encompassing 9 primary categories and numerous subcategories. Each instance contains a natural language query, a database schema, and the corresponding SQL query, accompanied by detailed annotations for the erroneous SQL queries. This dataset can be utilized in the fields of natural language processing and database management, to advance the practical deployment and robustness enhancement of NL2SQL technologies.




