CppSATD: Large-Scale Technical Debt Dataset for C++ Projects
收藏Zenodo2025-11-04 更新2026-05-26 收录
下载链接：
https://zenodo.org/doi/10.5281/zenodo.17523466
下载链接
链接失效反馈官方服务：
资源简介：
This is a comprehensive dataset that contains 531,398 annotated code comments for Self-Admitted Technical Debt Detection in C++ projects.
This dataset includes:- 531,398 total comments- 47 open source C++ projects- 6 annotation categories: DESIGN, DEFECT, DOCUMENTATION, REQUIREMENT, TEST, NON-SATD- Manual annotation with high inter-annotator agreement (κ > 0.75)Rich context: Code before and after the snippet
- File size: 197 MB (CSV format, UTF-8 encoded)
The dataset supports machine learning research in the detection of technical debt, and is associated with the paper "XAI-TD: Explainable AI for Technical Debt Detection in C++" by Saeed et al., 2025.Benchmark results using this dataset : - Random Forest: F1-score 0.914, Precision 0.999, Recall 0.842 - Dataset enables reproducible research in software maintenance and technical debt management
提供机构：
Zenodo创建时间：
2025-11-04
