We propose a new Graph Embedding (GEM) method that takes advantages of structural pattern representation. It models an Attributed Graph (AG) as a Probabilistic Graphical Model (PGM). Then, it learns the parameters of this PGM presented by a vector. This vector is a signature of AG in a lower dimensional vectorial space. We apply Structured Support Vector Machines (SSVM) to process classification task. As first tentative, results on the GREC dataset are encouraging enough to go further on this direction.
CITATION STYLE
Jarraya, H., Terrades, O. R., & Lladós, J. (2017). Graph embedding through probabilistic graphical model applied to symbolic graphs. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10255 LNCS, pp. 392–399). Springer Verlag. https://doi.org/10.1007/978-3-319-58838-4_43
Mendeley helps you to discover research relevant for your work.