EXPERT: Public Benchmarks for Dynamic Heterogeneous Academic Graphs
This provides new public benchmarks for researchers working on dynamic graph learning, addressing a gap in existing static and homogeneous datasets.
The authors tackled the lack of dynamic heterogeneous graph benchmarks by introducing large-scale datasets from AI and nuclear nonproliferation publications, and proposed an improved evaluation approach for graph forecasting models.
Machine learning models that learn from dynamic graphs face nontrivial challenges in learning and inference as both nodes and edges change over time. The existing large-scale graph benchmark datasets that are widely used by the community primarily focus on homogeneous node and edge attributes and are static. In this work, we present a variety of large scale, dynamic heterogeneous academic graphs to test the effectiveness of models developed for multi-step graph forecasting tasks. Our novel datasets cover both context and content information extracted from scientific publications across two communities: Artificial Intelligence (AI) and Nuclear Nonproliferation (NN). In addition, we propose a systematic approach to improve the existing evaluation procedures used in the graph forecasting models.