cat-edge-MAG-10 dataset
The MAG-10 network is a subset of the Microsoft Academic Graph where nodes are authors, hyperedges correspond to a publication from those authors, and the hyperedges are categorized by one of 10 computer science confereneces (e.g., "WWW", "KDD", "ICML"). If the same set of authors published at more than one conference, we used the most common venue as the category and any cases where there is a tie were discarded. Papers with more than 25 authors were also omitted. Some summary statistics of the network are:
  • number of nodes: 80,198
  • number of hyperedges: 51,889
  • number of edge label categories: 10
  • rank of hypergraph network (maximum hyperedge size): 25
Data files: If you use this data, please cite the following paper:
  • Clustering in graphs and hypergraphs with categorical edge labels.
    Ilya Amburg, Nate Veldt, and Austin R. Benson.
    Proceedings of WWW, 2020.
    [bibtex]
  • An overview of Microsoft Academic Service (MAS) and applications.
    Arnab Sinha, Zhihong Shen, Yang Song, Hao Ma, Darrin Eide, Bo-June Hsu, and Kuansan Wang.
    Proceedings of WWW, 2015.
    [bibtex]