Conceptual Modeling and Machine Learning: Opportunities, Challenges, and Lessons Learned
Conventionally, conceptual modeling is executed by humans, who employ abstraction to conceptualize segments of a reality or domain for the purposes of understanding and communication, and processing by machines. Recently, a trend toward integrating conceptual modeling and artificial intelligence, particularly machine learning, has emerged. This keynote illuminates the state of the art in this hybrid research field. The discussion will encompass both the significant opportunities and the fundamental challenges that must be addressed to fully exploit the potential of AI and ML for conceptual modeling.The presentation will include an overview of our recent research, focusing on i) the transformation of conceptual models into structures that facilitate the application of ML, and ii) the utilization of Large Language Models, Knowledge Graphs, and Graph Neural Networks to enable AI-assisted conceptual modeling.The conclusion of the talk will be devoted to the synthesis of lessons learned and the proposal of an agenda for future research directions.
Dominik Bork is an Associate Professor of Business Systems Engineering at the Faculty of Informatics, Institute of Information Systems Engineering, Business Informatics Group at TU Wien. His research interests comprise conceptual modeling, model-driven engineering, and modeling tool development. A primary focus of ongoing research is on the mutual benefits of conceptual modeling and artificial intelligence. More information can be found at https://www.model-engineering.info/.