Workflow process mining based on machine learning

This paper presents an algorithm of workflow process mining based on machine learning from the logs of business process instances, which can handle concurrence and recurrence of the business process that are the restrictions of other algorithms. Moreover workflow modeling language named flexible workflow modeling language (FWF-NET) is put forward, which can model uncertain and incomplete business process information, So the business process mined according to the algorithm can easily be transformed the FWF-NET. The prototype and experiments have proved that the algorithm mines business process effective, reduces the complexity in workflow process modeling and evolution, and evaluates performance of existing workflow model.