In this talk, I will introduce two recent studies -- one on graph ?edit distance (GED) computation and the other on community search, ?both using graph neural? networks.

For the first problem, we propose a novel deep learning framework ?that solves the GED problem in a two-step manner: 1) The proposed ?graph neural network GEDGNN is in charge of predicting the GED value ?and a matching matrix; and 2) A post-processing algorithm is used to ?derive ? possible node matchings from the matching matrix generated by ?GEDGNN. The best matching will lead to a high-quality graph edit path.

For the second problem, we propose a Query Driven-GNN (QD-GNN) model, ?in which, we combine the local query-dependent structure and global ?graph embedding to solve the community search problem.