A Systematic Literature Review of Machine Learning Applications in Software Engineering

Machine Learning (ML) has been a concern in Software Engineering (SE) over the past years. However, how to use ML and what it can offer for SE is still subject to debate among researchers. This paper investigates the application of ML in SE. The goal is to identify the used algorithms, the addressed topics and the main findings. It performs a Systematic Literature Review (SLR) of peer-reviewed studies published between 1995 and 2020. Data extracted from the studies show that ML algorithms are of great practical value in the different activities of software development process, especially “Software specification” and “Software validation” since “Software bug prediction” and “Software quality improvement” are the most recurring research topics. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.