Appendix for Activity Instance Identification using Bipartite Graph Matching

Overview

This appendix provides supplementary information to support the content and findings presented in the main paper, "Activity Instance Identification using Bipartite Graph Matching". Here, we offer additional data, and an exemplary implementation of our approach in order to enhance the understanding of the study's results.

Table of Contents

1. Appendix
2. Code
3. Datasets
4. References

Appendix

In this section, we present additional results (e.g. execution times, number of identified concept instances wrt all of the event logs) that prove the efficiency and the flexibility of our approach.

Contents:

• Evaluation
• The synthetic process model we use for simulation as a BPMN

Code

Any code or algorithms developed and used in this study are provided here. This includes scripts for the identification of concept instances. We also include instructions on how to execute these codes to reproduce our results.

Datasets

Contents:

• Preprocessed event log for BPI Challenge 2012
• Preprocessed event log for BPI Challenge 2017
• The synthetic event log with different types on injected noise

References

Finally, we list all additional references cited in this appendix. These references provide further context and support for the materials presented here.

Contents:

• Bernard, G., Andritsos, P.: Selecting representative sample traces from large event logs. In: ICPM (2021)
• van Dongen, B.: BPI challenge 2012 (2012)
• van Dongen, B.: BPI challenge 2017 (2017)
• Pele, O., Werman, M.: Fast and robust earth mover’s distances. In: ICCV (2009)

We hope that this appendix serves as a valuable resource for readers seeking a deeper understanding of our approach. Should you have any questions or require further information, please do not hesitate to contact us.

Thank you for your interest in our work.

Authors:

• Chiao-Yun Li
• Anton Antonov
• Wil M.P. van der Aalst