TITLE: Steps towards building smart modeling tools

ABSTRACT: The use of Machine Learning (ML) techniques to solve Model-Driven Engineering (MDE) problems has begun to gain traction. The use of ML techniques to enhance modeling tools by adding intelligent features may significantly improve the quality of the tools and the productivity of their users. This is an active line of work nowadays, but there are several open problems that need to be solved before it can really take off. In this talk, I will present an overview of the main obstacles that we have identified so far, as well as the approaches that we are implementing to address them. This will include topics like how to collect and organize thousands of modeling artifacts, how to build datasets, and the use of specific ML models to build interesting applications like recommender systems and model generators.

BIO (short): Jesús Sánchez Cuadrado is a Ramón y Cajal researcher at the Universidad de Murcia. Previously, he was an associate professor at the Universidad Autónoma de Madrid. His research has been focused on Model-Driven Engineering topics, notably model transformation languages and domain-specific languages. He has recently been working on topics related to the application of machine learning to modeling, including search engines and datasets, intelligent editors, and model generators. On these topics, he has created several tools which are available at https://models-lab.github.io/software/. For more information please visit http://sanchezcuadrado.es.