DeepGD: A Deep Learning Framework for Graph Drawing Using GNN

In the past decades, many graph drawing techniques have been proposed for generating aesthetically pleasing layouts. However, it remains a challenging task since different layout methods tend to highlight characteristics of graphs. Recently, studies on deep-learning-based algorithms emerged but they are often not generalizable arbitrary graphs without retraining. this article, we propose Convolutional-Graph-Neural-Network-based deep learning framework, DeepGD, which can draw once trained. It attempts generate layouts by compromising among multiple prespecified aesthetics considering good usually complies with simultaneously. order balance tradeoff, two adaptive training strategies, adjust weight factor each aesthetic dynamically during training. The quantitative and qualitative assessment DeepGD demonstrates that is capable effectively, while being flexible at accommodating criteria.