Evaluating the Quality of Discovered Process Models

In the domain of process mining the evaluation of models (i.e., “How can we measure the quality of a mined process model?”) is still subject to ongoing research. Because the types of models used in process mining are typically on a higher level of abstraction (they, for example, allow to capture concurrency), the problem of model evaluation is challenging. In this paper, we elaborate on the problem of process model evaluation, and we evaluate both new and existing fitness metrics for different levels of noise. The new metrics and the noise generation are based on Hidden Markov Models (HMMs).