Know Your Streams: On the Conceptualization, Characterization, and Generation of Intentional Event Streams
Authors: Andrea Maldonado, Christian Imenkamp, Hendrik Reiter, Thomas Seidl, Wilhelm Hasselbring, Martin Werner, Agnes Koschmider
Organizations: School of Engineering and Design, Technical University of Munich, Germany · Database Systems and Data Mining, Ludwig Maximilian University of Munich, Germany · Munich Center for Machine Learning, Munich, Germany · University of Bayreuth, Bayreuth, Germany · Christian-Albrechts-University Kiel, Kiel, Germany
The shift toward IoT-enabled, sensor-driven systems has transformed how operational data is generated, favoring continuous, real-time event streams (ES) over static event logs. This evolution presents new challenges for Streaming Process Mining (SPM), which must cope with out-of-order events, concurrent activities, incomplete cases, and concept drifts. Yet, the evaluation of SPM algorithms remains rooted in outdated practices, relying on static logs or artificially streamified data that fail to reflect the complexities of real-world streams. To address this gap, we first perform a comprehensive review of data stream literature to identify stream characteristics currently not reflected in the SPM community. Next, we use this information to extend the conceptual foundation for ES. Finally, we propose Stream of Intent, a prototype generator to produce ES with specific features. Our evaluation shows excellence in producing reproducible, intentional ES for targeted benchmarking and adaptive algorithm development in SPM.
Figures & tables
Figure 1: Know Your Streams is the first framework to breach ES research with the complexities of real-world real-time processes.
Figure 2: Comparing discovered process models (with dependency threshold=0.85 ) including vs. omitting realistic ES characteristics from the same event source.
Figure 3: Stream of Intent generates ES with intentional features, optimizing configuration parameters.
Figure 4: In the diagonal, we present ranges with low distance results; above the diagonal, average distances; and below it, distances comparing targets to generated event stream feature values for each combination of features.
Figure 5: Principal-component map contrasting generated ES (orange) with benchmark event logs (blue).