Leveraging multi-view deep learning for next activity prediction

Predicting the next activity in a running trace is a fundamental problem in business process monitoring since such predictive information may allow analysts to intervene proactively and prevent undesired behaviors. This paper describes a predictive process approach that couples multi-view learning and deep learning, in order to gain accuracy by accounting for the variety of information possibly recorded in event logs. Experiments with benchmark event logs show the accuracy of the proposed approach compared to several recent state-of-the-art methods.