Vectorial Signature for Matching Business Process Graphs

With the rapid proliferation of business process model collections, a fast and efficient systems to find process models from hundreds or thousands model is necessary. Similarity search or matching is one aspect needed to manage these collections, which consists to compare process models and select the fitted one to a query model. A graph based approaches for process model matching is very efficient in terms of accuracy. However, the complexity of graph matching algorithms grows exponentially with the size of graphs. In this paper, we propose an approach based on vectorial signatures to compare and evaluate the similarity of process models. To reduce the complexity of matching, we split the process graphs into paths, then calculate their similarity based on the similarity of the different vectorial signatures of their paths. Experimental evaluations show the performance of our approach using the precision and recall measures.