Process Mining and Machine Learning: Distributed Achievements, but Shared Challenges

Abstract

Process mining, as other diverse disciplines inspired by data mining, employs machine learning algorithms to enhance their functionality. Process mining solutions leverage the potential of machine learning to reduce difficulties in complex tasks. As a result, the proposal of innovative solutions has been boosted in tasks such as predicting the next event, monitoring process instances, or recommending resource allocation. However, dealing with machine learning in such a context poses special challenges: problem modeling, algorithm selection and configuration with attention to data (events) peculiarities, and the interpretation and usefulness of the results depend on the experience in business processes.