Authors: Markus Guentert, Mathias Weske, Matthias Kunze
Tags: 2012, conceptual modeling
With the increasing uptake of business process management efforts in companies, similarity search in large process model repositories has gained significance, as it forms a cornerstone of effective process model management and reuse. Similarity search uses a process model as query and retrieves all models, which resemble the query, in a ranked order. So far, the quality of the ranking has not been investigated. In this paper, we propose quality measures for similarity search results in order to address this problem, providing information on how good and how differentiated the results are. Our measures assess result statistics, which are derived from the similarity to the query model, and the agreement of different rankings, produced by diverse similarity measures. We apply our findings to a reference process model collection and comprehensively evaluate their prediction towards human assessment of process similarity.Read the full paper here: https://link.springer.com/chapter/10.1007/978-3-642-34002-4_17
