Stanford Web Graph Analyzed
A Stanford Web dataset of 2,312,497 hyperlinks, collected by Network Repository. Each record contains webpage-to-webpage hyperlinks. Stanford web graph from 2013. Suitable for web structure analysis, PageRank.
8 datasets tagged as the tag “Network repository”.
8 datasets.
A Stanford Web dataset of 2,312,497 hyperlinks, collected by Network Repository. Each record contains webpage-to-webpage hyperlinks. Stanford web graph from 2013. Suitable for web structure analysis, PageRank.
A RoadNet dataset of 1,921,660 roads, collected by Network Repository. Each record contains road intersections. Texas road network. Suitable for infrastructure analysis, network science.
A StackOverflow dataset of 1,301,942 edges, collected by Network Repository. Each record contains user-user interactions. StackOverflow user interaction network. Suitable for social network analysis, community detection.
The BookCrossing user–book network in its implicit form, as distributed by KONECT (bookcrossing_full-rating): 1,149,739 interactions between 105,278 members of the BookCrossing community and 340,523 books, collected by Cai-Nicolas Ziegler in August and September 2004. An edge records that a user interacted with a book; this version carries no rating values. Useful for collaborative filtering, implicit-feedback recommendation and the structure of user–item networks.
A Enron dataset of 1,148,072 emails, collected by Network Repository. Each record contains email communication. Enron email network from the 1999-2002 period. Suitable for organizational communication, network analysis.
A Facebook dataset of 817,035 edges, collected by Network Repository. Each record contains user-user friendships. Facebook social graph from WOSN. Suitable for social network analysis, community detection.
A Wikipedia dataset of 740,397 signed interactions, collected by Network Repository. Each record contains editor-to-editor signed relationships. Signed network of Wikipedia edit interactions (co-editing vs reverts). Suitable for signed network analysis, conflict detection, collaboration studies.
The citation network of the arXiv High Energy Physics – Phenomenology (hep-ph) section, from SNAP (the KDD Cup 2003 data), as distributed by KONECT (cit-HepPh): 421,578 citations among 34,546 papers posted between January 1993 and April 2003. An edge points from a citing paper to the paper it cites. Useful for citation analysis, scientific impact and the growth of networks over time.