TAR++: A New Process Model Similarity Algorithm Based on the Importance of TARs

摘要

As one of the three elements in enterprises, business process models are their very important asset. An effective process model similarity algorithm is a guarantee for process model retrieval, clustering and classification. In consideration of different deficiencies in existing algorithms, we present a new algorithm called TAR++ which is based on the importance of transition adjacent relations (TARs). The main idea of TAR++ is to describe the relationship of the transitions by adding an importance weight on TARs so as to generate a weighted TAR set and redefine union and intersection operators. In this paper, we introduce the TAR++ algorithm grounded in the Jaccard coefficient and leverages weighted TAR sets. The properties of TAR++ are validated on data sets of SAP, China CNR and DongFang Boiler Group.