Process Mining Influence on Requirement Elicitation for Machine Learning

Machine Learning (ML) is becoming a prominent approach in decision-making across many domains. However, the complexity of data and the high number of features are severe challenges when selecting the most effective features in ML models. This situation demands better quantity and quality requirements of the features that are the essential ingredients to ML models for higher accuracy. On the other hand, Process Mining is a young discipline which is successfully shown it's growth in process discovery using event logs. This study explores the influence of Process Mining by revealing quantitative and qualitative requirements for features to enhance the ML approaches. The proposed methodology was validated with a case of Colorectal Cancer (CRC) treatment planning.