Process model matching refers to the automatic identification of corresponding activities between two process models. It represents the basis for many advanced process model analysis techniques such as the identification of similar process parts or process model search. A central problem is how to evaluate the performance of process model matching techniques. Often, not even humans can agree on a set of correct correspondences. Current evaluation methods, however, require a binary gold standard, which clearly defines which correspondences are correct. The disadvantage of this evaluation method is that it does not take the true complexity of the matching problem into account and does not fairly assess the capabilities of a matching technique. In this paper, we propose a novel evaluation method for process model matching techniques. In particular, we build on the assessment of multiple annotators to define probabilistic notions of precision and recall. We use the dataset and the results of the Process Model Matching Contest 2015 to assess and compare our evaluation method. We find that our probabilistic evaluation method assigns different ranks to the matching techniques from the contest and allows to gain more detailed insights into their performance.
展开▼
