What is an Event Knowledge Graph?

An Event Knowledge Graph (EKG) is a graph-based data structure that stores events together with the objects involved and their relationships. Unlike a flat event log, an EKG can represent multiple process perspectives and their connections simultaneously, without reducing them to a single case ID.

Why does this matter?

Real processes rarely follow a single case notion. Event Knowledge Graphs overcome this limitation by connecting several dimensions at once – the basis for more realistic, multidimensional process analysis without artificial simplification.

What does this look like in practice?

An event log knows exactly one case ID, the order number for example. An event knowledge graph models order, delivery, invoice and customer as separate objects with relationships between them. A consolidated shipment covering three orders becomes one node with three edges — in a flat log the same event would have to be duplicated three times.

Not to be confused with

Not the same as a graph database, which is the storage technology; the event knowledge graph is the data model held inside it. Nor the same as a process model such as BPMN, which describes the intended flow — the graph holds the events actually observed and how they connect.