A machine and deep learning analysis among SonarQube rules, product, and process metrics for fault prediction

Abstract Background Developers spend more time fixing bugs refactoring the code to increase maintainability than developing new features. Researchers investigated quality impact on fault-proneness, focusing smells and metrics. Objective We aim at advancing fault-inducing commit prediction using different variables, such as SonarQube rules, product, process metrics, adopting techniques. Method designed conducted an empirical study among 29 Java projects analyzed with SZZ algorithm identify fault-fixing commits, computing product Moreover, we fault-proneness Machine Deep Learning models. Results 58,125 commits containing 33,865 faults infected by 174 rules violated 1.8M times, which 48 software metrics were calculated. clearly identified a set of features that provided highly accurate fault (more 95% AUC). Regarding performance classifiers, higher accuracy compared Conclusion Future works might investigate whether other static analysis tools, FindBugs or Checkstyle, can provide similar results. researchers consider adoption series anomaly detection