Detecting Sudden and Gradual Drifts in Business Processes from Execution Traces (Oct-2017)
The main aim of this project is to find the changes (drifts) in our business activities with the help of the execution traces.
In proposed System, automated and statistically grounded method for detecting sudden and gradual business process drifts under a unified framework. An evaluation over synthetic logs showed that the method accurately discovers typical process changes and nested compositions thereof, outperforming state-of-the art baselines both for sudden and gradual drifts. A separate evaluation on a large real-life log demonstrated the method’s ability to detect drifts that correspond to user-recognizable process changes, as well as its scalability.