Towards machine learning guided by best practices

Nowadays, machine learning (ML) is being used in software systems with multiple application fields, from medicine to software engineering (SE). On the one hand, the popularity of ML in the industry can be seen in the statistics showing its growth and adoption. On the other hand, its popularity can also be seen in research, particularly in SE, where multiple studies related to the use of Machine Learning in Software Engineering have been published in conferences and journals. At the same time, researchers and practitioners have shown that machine learning has some particular challenges and pitfalls. In particular, research has shown that ML-enabled systems have a different development process than traditional software, which also describes some of the challenges of ML applications. In order to mitigate some of the identified challenges and pitfalls, white and gray literature has proposed a set of recommendations based on their own experiences and focused on their domain (e.g., biomechanics), but for the best of our knowledge, there is no guideline focused on the SE community. This thesis aims to reduce the gap of not having clear guidelines in the SE community by using possible sources of practices such as question-and- answer communities and also previous research studies. As a result, we will present a set of practices with an SE perspective, for researchers and practitioners, including a tool for searching them.