Abstract
Graph neural network(GNN) models have been widely used in graph-structured data which can be seen everywhere but difficult to deal with. Link prediction is a kind of typical graph learning task which has been widely used in many applications such as knowledge graph completion, social recommendation, etc. Most existing link prediction models need preprocessing to preserve properties of graph such as asymmetric transitivity, node proximity, etc., which make models inconvenient to use. In this paper, we focus on link prediction in directed graphs and propose an end-to-end model called directed graph variational auto-encoder(DGVAE). DGVAE uses variational auto-encoder(VAE) as the whole framework and uses a two-layer GNN model FDGCN as the encoder. We design a brand new decoder for asymmetric output. Experiments are conducted on several different datasets and existing state-of-the-art models. Results show that DGVAE has the best performance in most cases.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.