Identifying Concerns When Specifying Machine Learning-Enabled Systems: A Perspective-Based Approach

Engineering machine learning (ML)-enabled systems poses various challenges from both a theoretical and a practical side. This thesis presents PerSpecML, a perspective-based approach for specifying ML-enabled systems that helps practitioners identify which attributes are important to contribute to the overall system’s quality. We evaluated PerSpecML in three different contexts: (i) in academia, (ii) with industry representatives, and (iii) in two real industrial case studies. The results particularly revealed key components that would have been otherwise missed without using PerSpecML.