Swevels, A., Dijkman, R. M., & Fahland, D. (2023). Inferring Missing Entity Identifiers from Context Using Event Knowledge Graphs. In C. Di Francescomarino, A. Burattin, C. Janiesch, & S. Sadiq (Eds.), Business Process Management: 21st International Conference, BPM 2023, Utrecht, The Netherlands, September 11–15, 2023, Proceedings (pp. 180-197). (Lecture Notes in Computer Science (LNCS); Vol. 14159). Springer. https://doi.org/10.1007/978-3-031-41620-0_11
Abstract
Complete event data is essential to perform rich analysis. However, real-life systems might fail in recording the (correct) case identifiers the system has operated on, resulting in incomplete event data. We aim to infer missing case identifiers of events by considering the physical constraints of the process which previous work has failed to do. We extended Event Knowledge Graphs (EKGs) with concepts for context and rule-based inference. We use the extended EKGs to model event data in its physical context and define five inference rules to infer identifiers of physical objects in a process. We evaluate the effectiveness of the rules on data from the IC manufacturing industry using conformance checking. Initially, none of the traces were complete. Our method inferred a case identifier for 95% of the events resulting in 88% complete traces.