Low-Modeling of (smart) Software
Jordi Cabot, Luxembourg Institute of Science and Technology (LIST), Luxembourg
Process Mining for Interdisciplinary Research
Agnes Koschmider, University of Bayreuth, Germany
System Modeling with Hybrid Quantum Machine Learning
Wolfgang Maaß, Saarland University, Germany
Low-Modeling of (smart) Software
Jordi Cabot
Luxembourg Institute of Science and Technology (LIST)
Luxembourg
https://jordicabot.com/
![]()
Brief Bio
Jordi Cabot is an FNR Pearl Chair and the head of the Software Engineering RDI Unit at the Luxembourg Institute of Science and Technology. Previously, he has been an ICREA Research Professor at Internet Interdisciplinary Institute, a Visiting Professor at the Western Norway University of Applied Sciences, an Associate Professor at École des Mines de Nantes (France) on an Inria international chair, a post-doctoral fellow at the University of Toronto, a senior lecturer at UOC and a visiting scholar at the Politecnico di Milano. He received the BSc and PhD degrees in Computer Science from the Technical University of Catalonia.
Current research topics include software and systems modeling, pragmatic formal verification techniques, analysis of open source and open data communities and the role AI and natural language interfaces can play in software development (and vice versa). He has published over 200 articles in top conferences and journals in these areas. Apart from his scientific publications, he writes and blogs about all these topics in several blogs such as the Modeling Languages portal.
Beyond scientific publications, the results of our research are available as open-source tools or as part of transfer contracts.
Abstract
There is an increasing demand for building better software and building it faster. This explains the growing popularity of low-code platforms, a new reincarnation of previous model-based approaches for software development.
Low-code platforms aim to speed up the development process by generating most of the application code from the high-level models. But what about those models? Can we also partially generate them? And if so, from what?
In this talk we will talk about low-modeling, a new approach to infer models out of different types of available information and (un)structured data sources. The goal of low-modeling is not only to increase the productivity of developer teams but also to contribute to the democratization of software development enabling non-technical users to build their own applications beyond what no-code and template-based approaches offer.
As example, we will focus on how low-modeling facilitates the generation of smart software, software that combines and traditional and AI-based components both in the front-end and back-end.
Process Mining for Interdisciplinary Research
Agnes Koschmider
University of Bayreuth
Germany
Brief Bio
Agnes Koschmider is Full Professor of Business Informatics at the University of Bayreuth and has a leading position in the Business Informatics branch of the Fraunhofer FIT. From 2019 to 2022 Agnes Koschmider was professor of business informatics at the Computer Science Institute of the University of Kiel. She completed her PhD in 2007 and her habilitation in Applied Informatics in 2015 at Karlsruher Institute of Technology (KIT).
She researches methods for data-driven analysis and explanation of processes (process mining), based on artificial intelligence, and methods for predicting process behavior. She is also researching on methods for privacy-preserving analysis and minimizing the re-identification of process data. At the center of her research is process analytics: developing a pipeline to efficiently process the complete chain from raw data (time series, sensor event data, and video data) to process discovery. The applications of such a data pipelines can be found in many disciplines such as medicine, agricultural sciences, geology, geography, material sciences or marine sciences.
Abstract
Process mining allows discovering bottlenecks in processes and revealing the deviations between real-life processes and to-be one. Process mining usually focuses on processing discrete event data, typically at the business level. The increasing volume of data demands techniques that can identify cause-effects in the data. Process mining promises to discover valuable knowledge from different types of data in terms of identifying anomalies in the processes and even explaining new effects within the data. For this purpose, however, existing process mining techniques must be adapted in order to meet the requirements of the other disciplines.
The first part of my talk outlines new fields of application for process mining and summarizes requirements for new process mining techniques. The second part of my talk shows implementations of the processing of various data types from raw data to processes and to AR visualization.
System Modeling with Hybrid Quantum Machine Learning
Wolfgang Maaß
Saarland University
Germany
Brief Bio
Wolfgang Maaß (engl. Maass) is a full professor at Saarland University for Business Administration, esp. Business Informatics at the Faculty of Humanities and Economics, co-opted professor for Computer Science at the Faculty of Mathematics and Computer Science at Saarland University, and associate professor for Biomedical Informatics at Stony Brook University, School of Medicine, NY. He is also a scientific director and head of the Smart Service Engineering research area at the German Research Center for Artificial Intelligence (DFKI). His research focuses on data economics, conceptual modeling, data-driven decision-making, and the use of artificial intelligence and quantum technologies for innovative services. He received his Ph.D. in computer science from Saarland University and his habilitation in business administration from the University of St. Gallen, Switzerland. His doctoral studies were funded by the German Research Foundation (DFG).
Abstract
Machine learning is becoming an integral part of operational information systems. They require increasingly large amounts of data to train powerful machine learning models. In some cases, such as the financial industry, pharmaceutical development, and industrial manufacturing, that optimization is no longer solvable even with supercomputers. Quantum computing shows ways to transform partial optimization problems by reducing computational complexity into the realm of feasibility. Hybrid Quantum Machine Learning combines machine learning approaches with quantum computing. This is exemplified by Quantum Embedding using the simulation of manufacturing processes.