Subgraph Query Matching in Multi-Graphs Based on Node Embedding

This paper presents an efficient algorithm for matching subgraph queries in a multi-graph based on features-based indexing techniques. The KD-tree data structure represents these nodes’ features, while the set-trie index multi-edges to make effectively. vertex core number, triangle and degree are eight features’ main features. densest query graph is extracted proposed model consists of two phases. first phase’s idea that, graph, find density similar neighborhood graph. Then k-nearest obtain subgraph. second phase each layer mapping feature vector (Vertex Embedding), improves model. To reduce node-embedding size be with KD-tree, dimension reduction, principal component analysis (PCA) method used. Furthermore, symmetry-breaking conditions will remove redundancy generated pattern In both phases, filtering process applied minimize number candidate nodes initiate vertex. Finally, testing effect concatenation structural features (orbits features) meta-features (summary general, statistical, information-theoretic, etc.) signatures performance. tested over three real benchmarks, datasets, randomly datasets. results agree theoretical study random cliques Erdos experiments showed that time efficiency scalability acceptable.