Online concept drift detection, localization andcharacterization using trace clustering

• Rafael Gaspar de Sousa USP
• Sarajane Marques Peres USP

DOI: https://doi.org/10.5753/sbsi.2020.13122

Resumo

Most process mining techniques assume stationary processes and are not well equipped to deal with concept drift. Online detection, localization and characterization of concept drift in business processes can support process mining techniques and analysts to improve organizations flexibility and adaptability. In this research, we propose a method to detect, locate and characterize concept drift in an online setting using trace clustering. The hypothesis is that the method can benefit from the trace clustering capacity to simplify complex problems through grouping similar patterns. In preliminary experiments, trace clustering was performed in a windowing setting showing that concept drift can be detected by analyzing the variation of clustering over time.

Palavras-chave: Online concept drift detection, localization, characterization trace clustering

Referências

Appice, A. and Malerba, D. (2016). A co-training strategy for multiple view clustering in process mining. IEEE Transaction on Services Computing, 9(6):832–845.

Bose, R. P. J. C., van der Aalst, W. M. P., ˆ Zliobaité, I., and Pechenizkiy, M. (2011). Handling concept drift in process mining. In Proceedings of the 23rd International Conference on Advanced Information Systems Engineering, pages 391–405, London, UK. Springer-Verlag.

Gonçaalves, P. M., de Carvalho Santos, S. G., Barros, R. S., and Vieira, D. C. (2014). A comparative study on concept drift detectors. Expert Systems Application, 41(18):8144 – 8156.

Hompes, B. F. A., Buijs, J., van der Aalst, W., Dixit, P., and Buurman, J. (2015). Detecting change in processes using comparative trace clustering. In Proceedings of the 5th International Symposium on Data-driven Process Discovery and Analysis, CEUR Workshop Proceedings, pages 95–108. CEUR-WS.org.

Kumar, M., Thomas, L., and Basava, A. (2015). Capturing the sudden concept drift in process mining. In Proceedings of the International Workshop on Algorithms Theories for the Analysis of Event Data, volume 1371, pages 132–143, Brussels, Belgium. CEUR-WS.org.

Luengo, D. and Sep´ulveda, M. (2012). Applying clustering in process mining to find different versions of a business process that changes over time. In Business Process Management Workshops, pages 153–158, Berlin, Heidelberg. Springer Berlin Heidelberg.

Maaradji, A., Dumas, M., Rosa, M. L., and Ostovar, A. (2015a). Business process drift. A synthetic dataset of 72 event logs based on Loan Assessment process containing different types of process drift.

Maaradji, A., Dumas, M., Rosa, M. L., and Ostovar, A. (2015b). Fast and accurate business process drift detection. In Business Process Management, pages 406–422, Innsbruck, Austria. Springer International Publishing.

Maita, A. R. C., Martins, L., Paz, C. R. L., Rafferty, L., Hung, P. C. K., Peres, S. M., and Fantinato, M. (2017). A systematic mapping study of process mining. Enterprise Information Systems, 12:1–45.

Ostovar, A. (2019). Business process drift: Detection and characterization. PhD thesis, Queensland University of Technology, Brisbane, Australia.

Ostovar, A., Maaradji, A., Rosa, M. L., and ter Hofstede, A. (2017). Characterizing drift from event streams of business processes. In 29th International Conference on Advanced Information Systems Engineering, pages 210–228, Switzerland. Springer.

van der Aalst,W. M. P. (2016). Process Mining: Data Science in Action. Springer, Berlin, Heidelberg, 2nd edition.

Zheng, C., Wen, L., and Wang, J. (2017). Detecting process concept drifts from event logs. In On the Move to Meaningful Internet Systems OTM 2017 Conference, pages 524–542, Cham. Springer International Publishing.