Machine Learning for Software Engineering: A Tertiary Study

Machine learning (ML) techniques increase the effectiveness of software engineering (SE) lifecycle activities. We systematically collected, quality-assessed, summarized, and categorized 83 reviews in ML for SE published between 2009 and 2022, covering 6,117 primary studies. The SE areas most tackled with ML are software quality and testing, while human-centered areas appear more challenging for ML. We propose a number of ML for SE research challenges and actions, including conducting further empirical validation and industrial studies on ML, reconsidering deficient SE methods, documenting and automating data collection and pipeline processes, reexamining how industrial practitioners distribute their proprietary data, and implementing incremental ML approaches.

Automated summary

Machine learning techniques improve the effectiveness of software engineering lifecycle activities. We collected, assessed, summarized, and categorized 83 reviews on ML for SE published between 2009 and 2022, covering 6,117 primary studies. ML is most commonly applied in software quality and testing, while human-centered areas pose greater challenges. We propose various research challenges and actions for ML in SE, including further empirical validation and industrial studies, reconsideration of deficient SE methods, documentation and automation of data collection and pipeline processes, reexamination of proprietary data distribution by industrial practitioners, and implementation of incremental ML approaches.