Blogcatalog, Citeseer, Cora, Cornell, Flickr, Pubmed, Texas, UAI2010, Washington, Wisconsin, Email, Wiki, ACM, Amazon, DBLP, IMDB, Cellphone, DBLP, Dynamic_cora, highSchool, Java

Dataset Overview

Dataset Categories

Complex Networks

• Blogcatalog
• Citeseer
• Cora
• Cornell
• Flickr
• Pubmed
• Texas
• UAI2010
• Washington
• Wisconsin

Topological Networks

• Email
• Wiki

Multilayer Networks

• ACM
• Amazon
• DBLP
• IMDB

Dynamic Networks

• Cellphone
• DBLP
• Dynamic_cora
• highSchool
• Java

Overlapping Complex Networks

Dataset Format

• name: dataset name
• topo: topology structure, type csr_matrix
• attr: attribute information, type csr_matrix
• label: label information, type csr_matrix, an n×k matrix where 1 indicates membership in the community, otherwise 0

Example Dataset Filenames

• Fb_X: Facebook X
• mag_chem: Chemistry
• mag_cs: Computer Science
• mag_med: Medicine
• mag_end: Engineering

Dataset Characteristics

1. All datasets contain self‑loops for nodes.
2. Apart from the label matrix, all other matrices contain only 1s or 0s.