INCREMENTAL MINING OF PROCESSES WITH LOOPS.

Currently most researches in process mining focus on discovering a workflow model from an entire log. In fact, the process designers may have a partially built model and their prior knowledge is also valuable information to process mining. Besides, the large volume of log data makes the process mining a time-consuming job. It is better that the process mining be done in an incremental way. There are only a few methods that can incrementally mine a model, but they have their limitations such as incapability in handling loops or being intolerant to noise. On the other hand, loop mining is a challenging problem in process mining because the repeatedly executed tasks add complexity to the search for task precedence. This paper studies the problem of handling loops in process mining and proposes an improved incremental process mining method which supports loops. Experiments in the end show the feasibility and validity of the proposed method. [ABSTRACT FROM AUTHOR]