Eindhoven University of Technology
Given an event log describing observed behavior, process discovery aims to find a process model that 'Best' describes this behavior. A large variety of process discovery algorithms has been proposed. However, no existing algorithm returns a sound model in all cases (free of deadlocks and other anomalies), handles infrequent behavior well and finishes quickly. The authors present a technique able to cope with infrequent behavior and large event logs, while ensuring soundness. The technique has been implemented in ProM and they compare the technique with existing approaches in terms of quality and performance.