Home // DATA ANALYTICS 2016, The Fifth International Conference on Data Analytics // View article

Improving Process Mining Prediction Results in Processes that Change over Time

Authors:
Alessandro Berti

Keywords: Concept Drift; Process Mining; Prediction.

Abstract:
In this paper, we propose a method in order to improve the accuracy of predictions, related to incomplete traces, in event logs that record changes in the underlying process. These ``second-order dynamics'' hamper the functioning of Process Mining discovery algorithms, but also hamper prediction results. The method is simple to implement as it is based exclusively on the Control Flow perspective and is computationally efficient. The approach has been validated on the Business Process Intelligence Challenge 2015's Municipality 5 event log, that contains an interesting shift in the process due to the union of the municipality with another municipality.

Pages: 37 to 42

Copyright: Copyright (c) IARIA, 2016

Publication date: October 9, 2016

Published in: conference

ISSN: 2308-4464

ISBN: 978-1-61208-510-4

Location: Venice, Italy

Dates: from October 9, 2016 to October 13, 2016