RegMiner
收藏资源简介:
RegMiner是由复旦大学开发的一个大型回归数据集,专注于从代码演化历史中自动收集可复现的回归错误。该数据集通过工具RegMiner在8周内从147个项目中收集了1035个回归错误,是目前已知最大的可复现回归数据集。数据集内容丰富,涵盖了多种软件工程和程序语言研究领域,如故障定位、软件测试和程序修复等。创建过程自动化,无需人工干预,确保了数据集的高精度和可接受召回率。应用领域广泛,旨在支持数据驱动的研究,为软件工程和程序语言社区提供了丰富的研究机会。
RegMiner is a large-scale regression dataset developed by Fudan University, which focuses on automatically collecting reproducible regression bugs from code evolution histories. Collected from 147 projects over 8 weeks using the RegMiner tool, this dataset contains 1035 regression bugs, making it the largest known reproducible regression dataset to date. With rich content, the dataset covers multiple research fields in software engineering and programming languages, such as fault localization, software testing, program repair, and more. Its creation process is fully automated without human intervention, ensuring high accuracy and acceptable recall rate of the dataset. It has a wide range of application scenarios, aiming to support data-driven research and provide abundant research opportunities for the software engineering and programming languages communities.




