Construction of Enterprise Architecture Knowledge Graphs (Extended Abstract) Philipp-Lorenz Glaser philipp-lorenz.glaser@tuwien.ac.at 0 Syed Juned Ali syed.juned.ali@tuwien.ac.at 0 Emanuel Sallinger emanuel.sallinger@tuwien.ac.at 1 Dominik Bork 0 Model-based EA Knowledge Graph Construction Enterprise Architecture, Knowledge Graph, Modeling Tool, ArchiMate, Archi Business Informatics Group , TU Wien , Austria Database and Artificial Intelligence Group , TU Wien , Austria

In this extended abstract, we report on a method for the model-based construction of Enterprise Architecture Knowledge Graphs (EAKG) and it's tool support by means of a Archi toolkit plugin. We briefly elaborate on the need for such an approach and tooling, and show, how an EAKG can be constructed by initially transforming an EA model into a a graph structure that is subsequently enriched by EA-specific, graph characteristics-specific, and EA Smell-specific knowledge.

 Abstract) -

organize and represent knowledge to ease advanced reasoning (e.g., rule-based and machine learning-based) and provide question-answering, recommendation, and information retrieval solutions. Therefore, we provide a knowledge graph-based generic EA analysis tool to utilize the full potential of graph-based representation of EA models.

In the full paper [ 4 ], we propose the model-based construction and enrichment of EAKGs to exploit the benefits of KG-based representation and reasoning in EA. The model-based construction involves three phases. Initially, knowledge is extracted and created from the EA models in the KG Creation phase. Then the knowledge is enriched, and additional inferences are made using that knowledge in the KG Enrichment phase. Fig. 1 visualizes the knowledge creation phase (left) and the Archi plugin we developed (right) (cf.[ 5 ]) where the EAKG is used to identify and visualize EA Smells [ 6 ].

Acknowledgements This research has been partly funded by the Austrian Research Promotion Agency via the Austrian Competence Center for Digital Production under the contract number 854187.

 [1] M. Smajevic , D. Bork , Towards graph-based analysis of enterprise architecture models , in: Int. Conf. on Conceptual Modeling , 2021 , pp. 199 - 209 . [2] D. Naranjo , M. Sánchez , J. Villalobos , Primrose: a graph-based approach for enterprise architecture analysis , in: Int. Conf. on Enterprise Information Systems , 2014 , pp. 434 - 452 . [3] A. Santana , D. Simon , K. Fischbach , H. de Moura, Combining network measures and expert knowledge to analyze enterprise architecture at the component level , in: 2016 IEEE EDOC Conference , IEEE, 2016 , pp. 1 - 10 . [4] P. Glaser , S. J. Ali , E. Sallinger , D. Bork , Model-based construction of enterprise architecture knowledge graphs , in: EDOC 2022 , 2022 , pp. 57 - 73 . doi: 10 .1007/978- 3- 031 - 17604- 3\_4. [5] P. Glaser , S. J. Ali , E. Sallinger , D. Bork , Exploring enterprise architecture knowledge graphs in archi: The EAKG toolkit , in: EDOC 2022 Workshops , 2022 , pp. 332 - 338 . [6] M. Smajevic , S. Hacks , D. Bork , Using knowledge graphs to detect enterprise architecture smells , in: The Practice of Enterprise Modeling - PoEM 2021 , 2021 , pp. 48 - 63 .