Detection of Sequences with Anomalous Behavior in a Workflow Process
A workflow process consists of an organized and repeatable pattern of activities that are necessary to complete a task, within the dynamics of an organization. The automatic recognition of deviations from the expected behavior within the workflow of an organization is crucial to provide assistance to new employees to accomplish his/her tasks. In this article, we propose a two-fold approach to this problem. First, taking the process logs as an input, we automatically build a statistical model that captures regularities in the activities carried out by the employees. Second, this model is used to track the activities performed by the employees to detect deviations from the expected behavior, according to the normal workflow of the organization. An experimental evaluation with five processes logs, with different levels of noise, was conducted to determine the validity of our approach.
KeywordsProcess mining Outliers detection
- 5.Ghionna, L., Greco, G., Guzzo, A., Pontieri, L.: Outlier detection techniques for process mining applications. In: An, A., Matwin, S., Ras, Z., Slezak, D. (eds.) Foundations of Intelligent Systems. Lecture Notes in Computer Science, vol. 4994, pp. 150–159. Springer, Heidelberg (2008)CrossRefGoogle Scholar
- 9.Hunter, J.S.: The exponentially weighted moving average. J. Qual. Technol. 18, 203–209 (1986)Google Scholar
- 10.Weijters, A.J.M.M., van der Aalst, W.M.P.: Rediscovering workflow models from event-based data using little thumb. Integr. Comput.-Aided Eng. 10, 151–162 (2003)Google Scholar
- 12.Toon Jouck, B.D.: Generating artificial event logs to compare process discovery techniques. In: Proceedings of the 4th International Symposium on Data-driven Process Discovery and Analysis (SIMPDA 2014), vol. 1293. CEUR Workshop Proceedings, Milan, Italy (2014)Google Scholar