BGU Social Networks Security Research Group has anonymized versions of directed (anybeat, Academia.edu, Google Plus), undirected (TheMarkerCafe), and multi-label (Students Network) social networks datasets that were crawled crawled by the group.
Memetracker data contains timestamped phrase and link information for news media articles and blog posts from 1.5 million different blogs and news websites.
The data spans 10 monts from August 2008 till May 2009, with several hundred million documents.
A set of sparse matrices from a wide spectrum of domains, including those arising from problems with underlying 2D or 3D geometry, like structural engineering, computational fluid dynamics, model reduction, electromagnetics, semiconductor devices, ...
A set of network datasets collected by Mark Newman.
Collection of networks for use in Pajek. Contains both small and larget networks and networks used at graph drawing competitions.
Set of networks from Center for Computational Analysis of Social and Organizational Systems (CASOS) at Carnegie Mellon University. Mostly small networks traditionally used in social science research.
Collection of networks from Trustlet. Mostly small online social networks with data on edges. The data usually denotes the level of trust or the rank of a user.
A collection of 2-mode networks between individual legislators and the bills with time stamps of link formation. The data was prepared by Yunkyu Sohn and James Fowler from UC San Diego.