A New Similarity Method based on Weighted Graph Models for Matching Parallel Business Process Models

Abstract

In business process, similarity is important for comparing between business process models. The existing similarity methods, such as Graph-based Matching Method (GMA), Weighted Graph Edit Distance (WGED), Weighted Node Adjacent Relation Similarity (WNARS), Tree Declarative Pattern Edit Distance (TPED) and Cosine-Tree Declarative Pattern (Cosine-TDP) can distinguish between AND, OR, and XOR relationships. However, they have drawbacks in detecting same relationships with different event logs. This paper proposes a new similarity method based on weighted graph models called weighted graph-based parallel process model matching (WGPPM) for computing the behaviour of parallel activity relationships. The proposed method utilizes the frequency of activity relationships as the weight in the graph model to measure the similarity between processes containing parallel relationships. WGPPM is compared with GMA, WGED, WNARS, TPED, and Cosine-TDP. The result shows that WGPPM is able to compute similarity between same parallel relationship with different event logs.

Focus : Function clone identification on cross‐platform

“…[6][7][8][9][10][11] These methods are usually inaccurate or their detection performance is limited by the efficiency of graph matching algorithms. 1,12 For instance, Genius 10 takes more than 1 week to construct their proposed codebook for only three software packages and the time complexity is quadratic in the number of training samples and linear in the cost of the bipartite graph matching algorithm.…”

“…[6][7][8][9][10][11] These methods are usually inaccurate or their detection performance is limited by the efficiency of graph matching algorithms. 1,12 For instance, Genius 10 takes more than 1 week to construct their proposed codebook for only three software packages and the time complexity is quadratic in the number of training samples and linear in the cost of the bipartite graph matching algorithm.Recently, with the rapid development of deep learning and graph neural network (GNN), 13,14 graph matching and classification problems such as social network graph analytics 15 and chemical formula matching 13 can be solved well by transforming a representative graph to an embedding.…”