Revisiting common bug prediction findings using effort-aware models

Authors:
Yasutaka Kamei;Shinsuke Matsumoto;Akito Monden;Ken-ichi Matsumoto;Bram Adams;Ahmed E. Hassan
Affiliations:
Software Analysis and Intelligence Lab (SAIL), School of Computing, Queen's University, Canada;Graduate School of Engineering, Kobe University, Japan;Graduate School of Information Science, Nara Institute of Science and Technology, Japan;Graduate School of Information Science, Nara Institute of Science and Technology, Japan;Software Analysis and Intelligence Lab (SAIL), School of Computing, Queen's University, Canada;Software Analysis and Intelligence Lab (SAIL), School of Computing, Queen's University, Canada
Venue:
ICSM '10 Proceedings of the 2010 IEEE International Conference on Software Maintenance
Year:
2010

Citing 0
Cited 8

High-impact defects: a study of breakage and surprise defects

Proceedings of the 19th ACM SIGSOFT symposium and the 13th European conference on Foundations of software engineering
Evaluating defect prediction approaches: a benchmark and an extensive comparison

Empirical Software Engineering
Bug prediction based on fine-grained module histories

Proceedings of the 34th International Conference on Software Engineering
Method-level bug prediction

Proceedings of the ACM-IEEE international symposium on Empirical software engineering and measurement
Modular construction of an analysis tool for mining software repositories

Proceedings of the 12th annual international conference companion on Aspect-oriented software development
Transfer defect learning

Proceedings of the 2013 International Conference on Software Engineering
Is lines of code a good measure of effort in effort-aware models?

Information and Software Technology
Studying software evolution using topic models

Science of Computer Programming

Quantified Score

Hi-index	0.00

Visualization

Abstract

Bug prediction models are often used to help allocate software quality assurance efforts (e.g. testing and code reviews). Mende and Koschke have recently proposed bug prediction models that are effort-aware. These models factor in the effort needed to review or test code when evaluating the effectiveness of prediction models, leading to more realistic performance evaluations. In this paper, we revisit two common findings in the bug prediction literature: 1) Process metrics (e.g., change history) outperform product metrics (e.g., LOC), 2) Package-level predictions outperform file-level predictions. Through a case study on three projects from the Eclipse Foundation, we find that the first finding holds when effort is considered, while the second finding does not hold. These findings validate the practical significance of prior findings in the bug prediction literature and encourage their adoption in practice.