Towards a Software Engineering Process for Developing Data-Driven Applications

Machine Learning and Artificial Intelligence allow the development of a new type of applications that automatically identify hidden patterns, process large amounts of data, and classify data according to aforementioned patterns. While they offer interesting solutions for several problems, they also impose challenges on software engineers in charge of structuring the development effort. The new applications require to incorporate additional specialists and their work into an overall development effort. We thus propose a software engineering process for data-driven applications.