Startseite // Forschung // FSTM // DCS // Forschungsgr... // Research Group on Methods and Tools for Software Engineering, DevOps and Artificial Intelligence

Research Group on Methods and Tools for Software Engineering, DevOps and Artificial Intelligence

Forschungsziele

General information

The MESSIR group is part of the LASSY laboratory. Our group focuses on methods and tools for Software Engineering, DevOps and Artificial Intelligence in order to improve the quality of IT systems. Our methods and tools are developed using sound scientific basis. We develop open source tools to support our languages and to allow for research collaboration or technology transfer with industrial partners. Our aim is to offer novel and efficient approaches for the engineers to ensure system development and deployment. Specific fields are currently under important development:

  • DevOps and Agile methods
  • Software engineering methods and tools for neural networks engineering
  • Software engineering methods and tools for ecological cyber physical systems

 

Highlights in 2019

The group has played a key role in the management of, and teaching support for the first and second-year students of the recently opened Bachelor in Computer Science (BiCS) at the University of Luxembourg. In this context, the BiCS Management Tool (BMT) has been improved by the team to ease the management of the projects students perform every semester along with either staff of the university or external collaborators.

Another highlight, was the the successful completion of the first BiCS Challenge, which had a two-fold goal: spread the voice about the BiCS while attracting motivated and talented high-school students to follow such an educational track. A number of industrial sponsors have supported this event by providing prices for the students.

Last but not least, the BicsLab, a R&D student laboratory has been setup with the supervision of a number of student semester projects around software, greenware, and senseware; a first industrial partnership agreement has been signed resulting in 2 semester projects in the company; a BiCS students voluntary cell has been started around positive IT solutions.

Three most interesting publications (or other achievements) in 2019 

  1. Jahic Benjamin; Guelfi Nicolas; Ries, Benoît. “Software Engineering for Dataset Augmentation using Generative Adversarial Networks”. Proceedings of the 10th IEEE International Conference on Software Engineering and Service Science. This paper presents a novel software engineering approach for dataset augmentation using neural networks. It proposes a rigorous process for generating synthetic data to improve the training of neural networks. An experimentation with the MNIST dataset has been performed which demonstrates a successful usage of our approach. Lastly, current issues are discussed.
  2. The first international workshop on Frontiers in Software Engineering Education (FISEE19 - https://www.laser-foundation.org/fisee/fisee-2019/) was co-organised with the LASER Foundation and colleagues from Innopolis University to discuss what education in Software Engineering needs, what should be changed and how new and traditional institutions can adapt to the fast pace of technology. The outcomes of this workshop are published as a volume of the Springer Lecture Notes in Computer Science (LNCS) to appear in 2020.
  3. Yasir Imtiaz Khan, Alexandros Konios, and Nicolas Guelfi. 2019. “A Survey of Petri Nets Slicing”. ACM Computing Surveys Volume 51, Issue 5 January 2019, 32 pages. In this article, different slicing techniques are studied along with their algorithms. A noteworthy use of this survey is for the selection and improvement of slicing techniques for optimizing the verification of state event models.

Mitglieder / Gast-Mitglieder