Object Centric Event

Object-centric event processing focuses on analyzing event data where events relate to multiple interacting objects, rather than a single case, offering a more realistic representation of complex processes. Current research emphasizes developing novel model architectures, such as graph neural networks and recurrent deep state-space models, to effectively handle the inherent complexity of these multi-object interactions and predict future events or process states. This approach improves the accuracy and interpretability of process mining, predictive process monitoring, and other applications by leveraging the richer information contained in object-centric event logs, leading to more insightful analyses of business processes and other dynamic systems.