Event Log Sampling for Predictive Monitoring

Lecture notes in business information processing, pp. 154–166

Abstract

Abstract Predictive process monitoring is a subfield of process mining that aims to estimate case or event features for running process instances. Such predictions are of significant interest to the process stakeholders. However, state-of-the-art methods for predictive monitoring require the training of complex machine learning models, which is often inefficient. This paper proposes an instance selection procedure that allows sampling training process instances for prediction models. We show that our sampling method allows for a significant increase of training speed for next activity prediction methods while maintaining reliable levels of prediction accuracy.

Authors 6

1. Mohammadreza Fani Sani corresponding Aachen Process and Data Science Chair

   RWTH Aachen University

   Affiliation as printed

   Process and Data Science Chair, RWTH Aachen University, Aachen, Germany

2. Mozhgan Vazifehdoostirani

   Eindhoven University of Technology

   Affiliation as printed

   Industrial Engineering and Innovation Science, Eindhoven University of Technology, Eindhoven, The Netherlands

3. Gyunam Park Aachen Process and Data Science Chair

   RWTH Aachen University

   Affiliation as printed

   Process and Data Science Chair, RWTH Aachen University, Aachen, Germany

4. Marco Pegoraro Aachen Process and Data Science Chair

   RWTH Aachen University

   Affiliation as printed

   Process and Data Science Chair, RWTH Aachen University, Aachen, Germany

5. Sebastiaan J. van Zelst Aachen Process and Data Science Chair

   Fraunhofer Institute for Applied Information Technology · RWTH Aachen University

   Affiliation as printed

   Fraunhofer FIT, Birlinghoven Castle, Sankt Augustin, Germany

   Process and Data Science Chair, RWTH Aachen University, Aachen, Germany

6. Wil M. P. van der Aalst Aachen Process and Data Science Chair

   Fraunhofer Institute for Applied Information Technology · RWTH Aachen University

   Affiliation as printed

   Fraunhofer FIT, Birlinghoven Castle, Sankt Augustin, Germany

   Process and Data Science Chair, RWTH Aachen University, Aachen, Germany