A new method for business process retrieval using breadth-first traversal

Due to the increasing of complex services in the deployed cloud platforms, business process retrieval has become a challenge for enterprises. However, traditional methods mostly focus on single attribute of business processes, rather than considering the attribute relationship and the loop structure. In this paper, we propose a breadth-first traversal (BFT) similarity calculation approach to overcome these drawbacks. First, we transform business process models into breadth-first traversal sequences (BFTS) using breadth-first traversal. Next, the BFTSs are divided into behavior sequences and structure sequences. Finally, the similarity values are calculated by weighting the two parts. The experimental results demonstrate our proposed method is accurate and time-saving in searching business processes.