Graph Neural Networks: A bibliometrics overview

Recently, graph neural networks (GNN) have become a hot topic in machine learning community. This paper presents Scopus-based bibliometric overview of the GNNs’ research since 2004 when GNN papers were first published. The study aims to evaluate trends, both quantitatively and qualitatively. We provide trend research, distribution subjects, active influential authors institutions, sources publications, most cited documents, topics. Our investigations reveal that frequent subject categories this field are computer science, engineering, telecommunications. In addition, source publications is Lecture Notes Computer Science. prolific or impactful institutions found United States, China, Canada. also must-read based on citation count future directions. analysis reveals node classification popular task, followed by link prediction, literature. Moreover, results suggest application convolutional attention mechanisms now among topics research. Finally, scalability, generalization, over-smoothing, explainability some directions pursue. • Graph Convolutional Networks mechanism Node Scalability,