IA4CM: An Intelligence Augmentation Approach for Designing Domain-Specific Conceptual Modeling Methods

Autor(en)

Alexander Voelz, Christine Strauss

Abstrakt

The design of Domain-Specific Modeling Methods (DSMMs) requires deep domain knowledge and careful structuring of metamodels, making the process complex and time-consuming. To address this challenge, we present IA4CM, an Intelligence Augmentation (IA) method that supports the creation of DSMMs by automating key aspects of metamodel design. Although IA4CM per se is domain agnostic, it enables the development of domain-specific methods by pro-viding intelligent recommendations for classes, attributes, relationships, and class hierarchies. IA4CM integrates a neuro-symbolic approach that learns from a repository of existing modeling methods to offer contextually relevant sugges-tions through the combination of knowledge engineering and IA techniques. The recommendations produced by this neuro-symbolic approach are contrasted with outputs from state-of-the-art Large Language Models (LLMs) to demonstrate the added value of domain specificity and structured reasoning in model design. The core functionalities include domain-specific recommendations for new classes and relationships, the identification of missing attributes, and the optimization of class hierarchies. By automating these tasks, IA4CM allows method engineers to focus on domain-specific requirements while improving the consistency and scalability of the resulting DSMMs. This approach reduces manual effort in metamodeling and supports the creation of domain-specific solutions.

Organisation(en)

Institut für Marketing und International Business

Band

II

Seiten

69 - 91

DOI

https://doi.org/10.1007/978-3-031-98660-4

Publikationsdatum

01-2026

Peer-reviewed

Ja

ÖFOS 2012

502050 Wirtschaftsinformatik, 102001 Artificial Intelligence

Link zum Portal

https://ucrisportal.univie.ac.at/de/publications/2d3ecf1f-4458-4e78-8636-6c7dc40f53da