Towards a Metadata Model for Process Data

Research in process mining is strongly data-driven and often relies on publicly available data sets for the evaluation and comparison of algorithms. However, the FAIR principles (findable, accessible, interoperable, and reusable research data) have not been widely adopted in this research area. To facilitate data documentation in this field, we propose an initial metadata model for process data called ProMeta. Our proposal is guided by existing metadata standards and metadata models, which have been proposed in related research areas. Additionally, we provide a validation of ProMeta based on existing documentations of research data published at 4TU.ResearchData and Zenodo. The proposed metadata model thereby aims to start a discussion among researchers in the community on how research data in process mining should be documented.