A Framework for Explainable Concept Drift Detection in Process Mining

Rapidly changing business environments expose companies to high levels of uncertainty. This uncertainty manifests itself in significant changes that tend occur over the lifetime a process and possibly affect its performance. It is important understand root causes such since this allows us react change or anticipate future changes. Research mining has so far only focused on detecting, locating characterizing not finding In paper, we aim close gap. We propose framework adds an explainability level onto concept drift detection provides insights into cause-effect relationships behind define different perspectives process, detect drifts these plug causality check determines whether can be causal each other. showcase effectiveness our by evaluating it both synthetic real event data. Our experiments show approach unravels novel executed processes.