Abstract
Artificial neural networks will always make a prediction, even when completely uncertain and regardless of the consequences. This obliviousness of uncertainty is a major obstacle towards their adoption in practice. Techniques exist, however, to estimate the two major types of uncertainty: model uncertainty and observation noise in the data. Bayesian neural networks are theoretically well-founded models that can learn the model uncertainty of their predictions. Minor modifications to these models and their loss functions allow learning the observation noise for individual samples as well. This paper is the first to apply these techniques to predictive process monitoring. We found that they contribute towards more accurate predictions and work quickly. However, their main benefit resides with the uncertainty estimates themselves that allow the separation of higher-quality from lower-quality predictions and the building of confidence intervals. This leads to many interesting applications, enables an earlier adoption of prediction systems with smaller datasets and fosters a better cooperation with humans.
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
Similar content being viewed by others
Notes
- 1.
https://data.4tu.nl (4TU Centre for Research Data).
- 2.
- 3.
- 4.
- 5.
A theoretical possibility of data leakage remains. In reality, some case variables such as “Amount” are possibly unknown at the beginning of the case, even though every event log has a value for them.
- 6.
References
van Dongen, B.F., Crooy, R.A., van der Aalst, W.M.P.: Cycle time prediction: when will this case finally be finished? In: Meersman, R., Tari, Z. (eds.) OTM 2008. LNCS, vol. 5331, pp. 319–336. Springer, Heidelberg (2008). https://doi.org/10.1007/978-3-540-88871-0_22
Van der Aalst, W.M.P., Schonenberg, M.H., Song, M.: Time prediction based on process mining. Inf. Syst. 36, 450–475 (2011)
Polato, M., Sperduti, A., Burattin, A., de Leoni, M.: Data-aware remaining time prediction of business process Instances. Presented at the (2014)
Tax, N., Verenich, I., La Rosa, M., Dumas, M.: Predictive business process monitoring with LSTM neural networks. Lecture Notes Computer Science, vol. 10253, pp. 477–492 (2017)
Navarin, N., Vincenzi, B., Polato, M., Sperduti, A.: LSTM networks for data-aware remaining time prediction of business process instances. arXiv:1711.03822v1 (2017)
Verenich, I., Dumas, M., La Rosa, M., Maggi, F.M., Teinemaa, I.: Survey and cross-benchmark comparison of remaining time prediction methods in business process monitoring. ACM Trans. Intell. Syst. Technol. (TIST) 10(4), 1–34 (2019)
Kendall, A., Gal, Y.: What uncertainties do we need in Bayesian deep learning for computer vision? Presented at the (2017)
MacKay, D.: Bayesian methods for neural networks: theory and applications. In: Neural Networks Summer School. University of Cambridge (1995)
Gal, Y.: Uncertainty in deep learning: PhD thesis. University of Cambridge (2016). http://mlg.eng.cam.ac.uk/yarin/thesis/thesis.pdf
Kingma, D.P., Welling, M.: Auto-encoding variational Bayes. arXiv:1312.6114 (2014)
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15, 1929–1958 (2014)
Gal, Y., Ghahramani, Z.: Dropout as a Bayesian approximation: representing model uncertainty in deep learning. In: Proceedings of the 33rd International Conference on Machine Learning, vol. 48 (2016)
Gal, Y., Hron, J., Kendall, A.: Concrete dropout. Presented at the (2017)
Maddison, C.J., Mnih, A., Teh, Y.W.: The concrete distribution: a continuous relaxation of discrete random variables arXiv:1611.00712v3 (2017)
Gal, Y., Ghahramani, Z.: Bayesian convolutional neural networks with Bernoulli approximate variational inference. arXiv:1506.02158v1 (2015)
Gal, Y., Ghahramani, Z.: A theoretically grounded application of dropout in recurrent neural networks. Presented at the (2016)
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)
LeCun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Presented at the (1998)
Bai, S.J., Kolter, J.Z., Koltun, V.: An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv:1803.01271v2 (2018)
Weytjens, H., De Weerdt, J.: Process outcome prediction: CNN vs. LSTM (with attention). Presented at the (2020)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Weytjens, H., De Weerdt, J. (2021). Learning Uncertainty with Artificial Neural Networks for Improved Remaining Time Prediction of Business Processes. In: Polyvyanyy, A., Wynn, M.T., Van Looy, A., Reichert, M. (eds) Business Process Management. BPM 2021. Lecture Notes in Computer Science(), vol 12875. Springer, Cham. https://doi.org/10.1007/978-3-030-85469-0_11
Download citation
DOI: https://doi.org/10.1007/978-3-030-85469-0_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-85468-3
Online ISBN: 978-3-030-85469-0
eBook Packages: Computer ScienceComputer Science (R0)