Research Artifact: Predicting Defective Lines Using a Model-Agnostic Technique

https://github.com/awsm-research/line-level-defect-prediction

This is a research artifact for the paper: Predicting Defective Lines Using a Model-Agnostic Technique. This artifact is a repository consisting of our benchmark file-level and line-level defect datasets and results of our file-level defect prediction models. The purposes of this artifact are to enable researchers to discuss the predictive performance of the file-level defect prediction models, and to reuse the dataset for further software engineering research.

Contents

• Dataset - our benchmark file-level and line-level defect datasets, including 32 releases of 9 software studied systems.
• File-level_results.csv - results of our file-level defect prediction models based on within-release and cross-release evaluation settings.
• LICENSE.md - CC0 1.0 Universal.
• README.md - this file.

Authors

• Supatsara Wattanakriengkrai
• Patanamon Thongtanunam
• Chakkrit Tantithamthavorn
• Hideaki Hata
• Kenichi Matsumoto