Efficient Method for Detecting and Localizing Concept Drifts from Event Logs In Process Mining

Trupti Kakkad, Dr. Rahila Sheikh

Abstract: Although most business processes modification over time, modern process mining techniques tend to survey these processes as if they are in a balanced state. The change in processes may be immediately or moderately. The drift could also be periodic (e. g. , due to seasonal influences) or one-of-a-kind (e. g. , the consequences of recent legislation). To find and perceive such idea drifts in processes is crucial for the process management. A generic framework and specific techniques to notice once a method changes and to localize the components of the method that have modified are given during this paper. To delineate relationships among activities, completely different options are projected. Variations between consecutive populations are discovered by victimization these options. Like as a plug-in of the promenade methodology mining framework, The approach has been implemented and has been evaluated using every simulated event data exhibiting controlled conception drifts and from a Dutch municipality, real-life event information.

Keywords: Concept drift, flexibility, hypothesis tests, process changes, process mining,

How to Cite?: Trupti Kakkad, Dr. Rahila Sheikh, "Efficient Method for Detecting and Localizing Concept Drifts from Event Logs In Process Mining", Volume 5 Issue 1, January 2016, International Journal of Science and Research (IJSR), Pages: 178-182, https://www.ijsr.net/getabstract.php?paperid=1011601, DOI: https://dx.doi.org/10.21275/1011601