The Koblenz Network Collection

KONECT is a project to collect large network datasets to support research in the area of network mining. KONECT has over 100 datasets from sources such as arXiv, Amazon, Digg, DBLP, Enron, Flickr, Twitter, and Youtuve. KONECT also provides code to generate network datasets from the Web.

---

KONECT (Koblenz Network Collection) is a project to collect large network datasets to support network mining research. It is led by the Institute of Web Science and Technologies of the University of Koblenz-Landau. KONECT contains over a hundred network datasets, in such categories as

• Authorship (astrophysics, DBLP, Wikipedia, ...)
• Co-occurence (Amazon)
• Communication (Digg, Enron, ...)
• Features (Flickr, LiveJournal, Orkut, ...)
• Interaction (Caenorhabditis elegans, Haggle, ...)
• Physical (California, Gnutella, ...)
• Ratings (Epinions, MovieLens, ...)
• Reference (CiteSeer, DBLP, ...)
• Semantic (DBpedia)
• Social (Facebook, Filmtipset, ...)
• Trust (Epinions, ...)

A network as provided by KONECT is a set of nodes connected by links. An example of a network is a social network: a set of users connected by links which represent friendship relations. A network is represented mathematically by a graph, in which nodes are called vertices and links are called edges.

KONECT provides:

• Code to generate all network datasets from the Web
• Statistics and plots viewable online
• Download of selected datasets (where legally possible)

Access KONECT at konect.uni-koblenz.de/