Subgraph matching is acrucial task in many fields, ranging from information retrieval, computervision, biology, chemistry and natural language processing . Yet subgraphmatching problem remains to be an NP-complete problem . Study proposes anend-to-end learning-based approximate method for subgraph matching task, calledsubgraph matching network (Sub-GMN) The proposed Sub-GMn firstly uses graphrepresentation learning to map nodes to node-level embedding . It then combinesmetric learning and attention mechanisms to model the relationship betweenmatched nodes in the data graph and query graph . On average running time Sub-GNN runs 20-40times faster than FGNN . In addition, the average F1-score of Sub-MHN on allexperiments with dataset 2 reached 0.95, which demonstrates that Sub-MGNoutputs more correct node-to.-node matches. It can output a list of node-To-node matches, while most existing GNNs-based methods cannot provide the matched node pairs. Another advantage of our proposedSub-MHNN is that it can output the list of matches, which can be found in the

Author(s) : Zixun Lan, Limin Yu, Linglong Yuan, Zili Wu, Fei Ma

Links : PDF - Abstract

Code :
Coursera

Keywords : node - matching - learning - gmn - subgraph -

Leave a Reply

Your email address will not be published. Required fields are marked *