Abstract
Deviation detection is a set of techniques that identify deviations from normative processes in real process executions. These diagnostics are used to derive recommendations for improving business processes. Existing detection techniques identify deviations either only on the process instance level or rely on a normative process model to locate deviating behavior on the event level. However, when normative models are not available, these techniques detect deviations against a less accurate model discovered from the actual behavior, resulting in incorrect diagnostics. In this paper, we propose a novel approach to detect deviation on the event level by identifying frequent common behavior and uncommon behavior among executed process instances, without discovering any normative model. The approach is implemented in ProM and was evaluated in a controlled setting with artificial logs and real-life logs. We compare our approach to existing approaches to investigate its possibilities and limitations. We show that in some cases, it is possible to detect deviating events without a model as accurately as against a given precise normative model.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
We omit \(G \) and \(G' \) for both \(\lambda \) and \(\overline{\lambda }\) where the context is clear.
- 2.
Both the plugins and the experiments can be found in the TraceMatching package of the ProM.
- 3.
In this paper, we only discuss the accuracy score. However, one may use the confusion matrix and compute the F1 score of event identification or swap the confusion matrix to compute the F1 score of deviation identification. We have computed all three, and they have shown similar results.
- 4.
Using filter from 0.0 to 0.2, IMinf returns a flower model which is the same as classifying all events as conforming.
- 5.
- 6.
For each case, we added one deviating event resulting in a log with 13.2 % deviating events. Repeating this five times, we show the average acc scores.
References
van der Aalst, W.M.P., Adriansyah, A., van Dongen, B.F.: Replaying history on process models for conformance checking and performance analysis. Wiley Interdisc. Rev. Data Min. Knowl. Disc. 2(2), 182–192 (2012)
Bose, R., van der Aalst, W.M.P.: Context aware trace clustering: towards improving process mining results. In: Proceedings of the SIAM International Conference on Data Mining, SDM 2009, Sparks, Nevada, USA, 30 April–2 May, pp. 401–412 (2009)
Bose, R.P.J.C., van der Aalst, W.M.P.: Process diagnostics using trace alignment: opportunities, issues, and challenges. Inf. Syst. 37(2), 117–141 (2012)
Carmona, J., Cortadella, J.: Process discovery algorithms using numerical abstract domains. IEEE Trans. Knowl. Data Eng. 26(12), 3064–3076 (2014)
De Weerdt, J., vanden Broucke, S.K.L.M., Vanthienen, J., Baesens, B.: Active trace clustering for improved process discovery. IEEE Trans. Knowl. Data Eng. 25(12), 2708–2720 (2013)
Fahland, D., van der Aalst, W.M.P.: Simplifying discovered process models in a controlled manner. Inf. Syst. 38(4), 585–605 (2013)
Ghionna, L., Greco, G., Guzzo, A., Pontieri, L.: Outlier detection techniques for process mining applications. In: An, A., Matwin, S., Raś, Z.W., Ślęzak, D. (eds.) ISMIS 2008. LNCS (LNAI), vol. 4994, pp. 150–159. Springer, Heidelberg (2008)
Leemans, S.J.J., Fahland, D., van der Aalst, W.M.P.: Discovering block-structured process models from event logs containing infrequent behaviour. In: Lohmann, N., Song, M., Wohed, P. (eds.) BPM 2013 Workshops. LNBIP, vol. 171, pp. 66–78. Springer, Heidelberg (2014)
Lu, X., Fahland, D., van der Aalst, W.M.P.: Conformance checking based on partially ordered event data. In: Fournier, F., Mendling, J. (eds.) BPM 2014 Workshops. LNBIP, vol. 202, pp. 75–88. Springer, Heidelberg (2015)
Nguyen, H., Dumas, M., La Rosa, M., Maggi, F.M., Suriadi, S.: Mining business process deviance: a quest for accuracy. In: Meersman, R., Panetto, H., Dillon, T., Missikoff, M., Liu, L., Pastor, O., Cuzzocrea, A., Sellis, T. (eds.) OTM 2014. LNCS, vol. 8841, pp. 436–445. Springer, Heidelberg (2014)
Rebuge, A., Ferreira, D.: Business process analysis in healthcare environments: a methodology based on process mining. Inf. Syst. 37(2), 99–116 (2012)
Suriadi, S., Wynn, M.T., Ouyang, C., ter Hofstede, A.H.M., van Dijk, N.J.: Understanding process behaviours in a large insurance company in Australia: a case study. In: Salinesi, C., Norrie, M.C., Pastor, Ó. (eds.) CAiSE 2013. LNCS, vol. 7908, pp. 449–464. Springer, Heidelberg (2013)
Yang, W., Hwang, S.: A process-mining framework for the detection of healthcare fraud and abuse. Expert Syst. Appl. 31(1), 56–68 (2006)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Lu, X., Fahland, D., van den Biggelaar, F.J.H.M., van der Aalst, W.M.P. (2016). Detecting Deviating Behaviors Without Models. In: Reichert, M., Reijers, H. (eds) Business Process Management Workshops. BPM 2016. Lecture Notes in Business Information Processing, vol 256. Springer, Cham. https://doi.org/10.1007/978-3-319-42887-1_11
Download citation
DOI: https://doi.org/10.1007/978-3-319-42887-1_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-42886-4
Online ISBN: 978-3-319-42887-1
eBook Packages: Computer ScienceComputer Science (R0)