PGB: A PubMed Graph Benchmark for Heterogeneous Network Representation Learning
This provides a standardized benchmark for researchers in biomedical informatics and graph learning to test heterogeneous network methods, though it is incremental as it aggregates existing data into a new dataset.
The authors tackled the lack of a benchmark for evaluating heterogeneous graph embeddings in biomedical literature by introducing PubMed Graph Benchmark (PGB), a dataset with rich metadata from over 33 million PubMed articles, and they made it publicly available for tasks like systematic reviews, node classification, and node clustering.
There has been rapid growth in biomedical literature, yet capturing the heterogeneity of the bibliographic information of these articles remains relatively understudied. Although graph mining research via heterogeneous graph neural networks has taken center stage, it remains unclear whether these approaches capture the heterogeneity of the PubMed database, a vast digital repository containing over 33 million articles. We introduce PubMed Graph Benchmark (PGB), a new benchmark dataset for evaluating heterogeneous graph embeddings for biomedical literature. The benchmark contains rich metadata including abstract, authors, citations, MeSH terms, MeSH hierarchy, and some other information. The benchmark contains three different evaluation tasks encompassing systematic reviews, node classification, and node clustering. In PGB, we aggregate the metadata associated with the biomedical articles from PubMed into a unified source and make the benchmark publicly available for any future works.