Description: This dataset is collected from DBLP-Citation-network V12, which contains around 4.9 million papers and 45 million citation relationships. We construct graphs by selecting authors as nodes and co-authorships as edges during the time period from 1990 to 2019. To cut the graphs into pieces, we generate sub-graphs based on the Fields of Study attribute from papers. For each field, we generate one spatio-temporal graph. We generate 2361 spatio-tempora graphs with a total of around 9 million nodes and a total of around 6 million of edges.
Statistics:
Name
Type
#Graphs
#Nodes
#Edges
Attributed
Directed
Weighted
Signed
Homogeneous
Spatial
Temporal
Labels
CollabNet
Collaboration Networks
2,361
303,308
207,632
NO
NO
NO
NO
YES
Lat,Long
YES
YES
Acknowlegement: GraphGT: Machine Learning Datasets for Graph Generation and Transformation. Yuanqi Du, Shiyu Wang Xiaojie Guo, et al, 2021.