Software Engineering Research

Software engineering research investigates how large, evolving codebases can be understood, measured, and systematically improved—examining everything from how duplicate code fragments spread through a system to how static attributes of source code predict where bugs are likely to emerge. Empirical studies in this area apply machine learning to problems like locating defects before they cause failures, tracing requirements through layers of documentation and implementation, and detecting when refactoring has genuinely improved a system's structure rather than just rearranged it. A central open question is how well findings from controlled studies transfer to the messy, heterogeneous conditions of real industrial software. Researchers are also working to close the gap between automated analysis tools and the tacit knowledge developers rely on when deciding which warnings to act on and which to ignore.

Works

106,948

Total citations

1,578,769

Keywords

RefactoringCode Clone DetectionSoftware Defect PredictionRequirements TraceabilityStatic Code AttributesMachine Learning