Abstract
Deep learning models for next event prediction in predictive process monitoring have shown significant performance improvements over conventional methods. However, they are often criticized for being black-box models. Without allowing analysts to understand what such models have learned, it is difficult to establish trust in their abilities.
In this work, we propose a technique to infer a likelihood graph from a next event predictor (NEP) to capture and visualize its behavior. Our approach first generates complete cases, including event attributes, using the NEP. From this set of cases, a multi-perspective likelihood graph is inferred. Including event attributes in the graph allows analysts to better understand the learned decision and branching points of the process.
The results of the evaluation show that inferred graphs generalize beyond the event log, achieve high F-scores, and small likelihood deviations. We conclude black-box NEP can be used to generate conforming cases even for noisy event logs. As a result, our visualization technique, which represents exactly this set of cases, shows what the NEP has learned, thus mitigating one of their biggest criticisms.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
van Dongen, Boudewijn (2020): BPI Challenge 2020. 4TU.ResearchData. Collection. https://doi.org/10.4121/uuid:52fb97d4-4588-43c9-9d04-3604d4613b51.
References
Barredo Arrieta, A., et al.: Explainable artificial intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI. Inf. Fusion 58, 82–115 (2020)
Camargo, M., Dumas, M., González-Rojas, O.: Learning accurate LSTM models of business processes. In: Hildebrandt, T., van Dongen, B.F., Röglinger, M., Mendling, J. (eds.) BPM 2019. LNCS, vol. 11675, pp. 286–302. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-26619-6_19
Evermann, J., Rehse, J.R., Fettke, P.: Predicting process behaviour using deep learning. Decis. Support Syst. 100, 129–140 (2017)
Gal, A., Senderovich, A.: Process minding: closing the big data gap. In: Fahland, D., Ghidini, C., Becker, J., Dumas, M. (eds.) BPM 2020. LNCS, vol. 12168, pp. 3–16. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58666-9_1
Galanti, R., Coma-Puig, B., de Leoni, M., Carmona, J., Navarin, N.: Explainable predictive process monitoring. In: International Conference on Process Mining (2020)
Mehdiyev, N., Evermann, J., Fettke, P.: A novel business process prediction model using a deep learning method. Bus. Inf. Syst. Eng. 62(2), 143–157 (2018). https://doi.org/10.1007/s12599-018-0551-3
Nolle, T., Luettgen, S., Seeliger, A., Mühlhäuser, M.: BINet: multi-perspective business process anomaly classification. In: Information Systems (2019)
Nolle, T., Seeliger, A., Mühlhäuser, M.: BINet: multivariate business process anomaly detection using deep learning. In: Weske, M., Montali, M., Weber, I., vom Brocke, J. (eds.) BPM 2018. LNCS, vol. 11080, pp. 271–287. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-98648-7_16
Peeperkorn, J., Vanden Broucke, S.K.L.M., De Weerdt, J.: Can deep neural networks learn process model structure ? An assessment framework and analysis. In: International Workshop on Leveraging Machine Learning in Process Mining (2021)
Polyvyanyy, A., Moffat, A., García-Bañuelos, L.: An entropic relevance measure for stochastic conformance checking in process mining. In: International Conference on Process Mining, pp. 97–104 (2020)
Rehse, J.-R., Mehdiyev, N., Fettke, P.: Towards explainable process predictions for industry 4.0 in the DFKI-Smart-Lego-Factory. KI - Künstliche Intelligenz 33(2), 181–187 (2019). https://doi.org/10.1007/s13218-019-00586-1
Rizzi, W., Di Francescomarino, C., Maggi, F.M.: Explainability in predictive process monitoring: when understanding helps improving. In: Fahland, D., Ghidini, C., Becker, J., Dumas, M. (eds.) BPM 2020. LNBIP, vol. 392, pp. 141–158. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58638-6_9
Shunin, T., Zubkova, N., Shershakov, S.: Neural approach to the discovery problem in process mining. In: Analysis of Images, Social Networks and Texts, pp. 261–273 (2018)
Sindhgatta, R., Moreira, C., Ouyang, C., Barros, A.: Exploring interpretable predictive models for business processes. In: Fahland, D., Ghidini, C., Becker, J., Dumas, M. (eds.) BPM 2020. LNCS, vol. 12168, pp. 257–272. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58666-9_15
Sindhgatta, R., Ouyang, C., Moreira, C., Liao, Y.: Interpreting predictive process monitoring benchmarks. arXiv (2019)
Sommers, D., Menkovski, V., Fahland, D.: Process discovery using graph neural networks. In: International Conference on Process Mining, pp. 40–47 (2021)
Tax, N., Verenich, I., La Rosa, M., Dumas, M.: Predictive business process monitoring with LSTM neural networks. In: Dubois, E., Pohl, K. (eds.) CAiSE 2017. LNCS, vol. 10253, pp. 477–492. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-59536-8_30
Tax, N., van Zelst, S.J., Teinemaa, I.: An experimental evaluation of the generalizing capabilities of process discovery techniques and black-box sequence models. In: Gulden, J., Reinhartz-Berger, I., Schmidt, R., Guerreiro, S., Guédria, W., Bera, P. (eds.) BPMDS/EMMSAD -2018. LNBIP, vol. 318, pp. 165–180. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-91704-7_11
Theis, J., Darabi, H.: Decay replay mining to predict next process events. IEEE Access 7, 119787–119803 (2019)
van der Aalst, W.M.P.: Process Mining: Data Science in Action. Springer, Heidelberg (2016). https://doi.org/10.1007/978-3-662-49851-4
Verenich, I., Dumas, M., La Rosa, M., Nguyen, H.: Predicting process performance: a white-box approach based on process models. J. Softw. Evol. Process (2019)
Weinzierl, S., et al.: An empirical comparison of deep-neural-network architectures for next activity prediction using context-enriched process event logs. arXiv (2020)
Weinzierl, S., Zilker, S., Brunk, J., Revoredo, K., Matzner, M., Becker, J.: XNAP: making LSTM-based next activity predictions explainable by using LRP. In: Del Río Ortega, A., Leopold, H., Santoro, F.M. (eds.) BPM 2020. LNBIP, vol. 397, pp. 129–141. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-66498-5_10
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Gerlach, Y., Seeliger, A., Nolle, T., Mühlhäuser, M. (2022). Inferring a Multi-perspective Likelihood Graph from Black-Box Next Event Predictors. In: Franch, X., Poels, G., Gailly, F., Snoeck, M. (eds) Advanced Information Systems Engineering. CAiSE 2022. Lecture Notes in Computer Science, vol 13295. Springer, Cham. https://doi.org/10.1007/978-3-031-07472-1_2
Download citation
DOI: https://doi.org/10.1007/978-3-031-07472-1_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-07471-4
Online ISBN: 978-3-031-07472-1
eBook Packages: Computer ScienceComputer Science (R0)