Cognifying Model-Driven Software Engineering

Abstract

The limited adoption of Model-Driven Software Engineering (MDSE) is due to a variety of social and technical factors, which can be summarized in one: its (real or perceived) benefits do not outweigh its costs. In this vision paper we argue that the cognification of MDSE has the potential to reverse this situation. Cognification is the application of knowledge (inferred from large volumes of information, artificial intelligence or collective intelligence) to boost the performance and impact of a process. We discuss the opportunities and challenges of cognifying MDSE tasks and we describe some potential scenarios where cognification can bring quantifiable and perceivable advantages. And conversely, we also discuss how MDSE techniques themselves can help in the improvement of AI, Machine learning, bot generation and other cognification techniques.