Project: KONECT-Cloud Large-scale Network Mining in the Cloud

Team: Prof. Dr. Steffen Staab, Dr. Dr. Sergej Sizov, Dr. Thomas Gottron, Dr. Jérôme Kunegis

Research institution: Universität Koblenz-Landau

Abstract: In the proposed project, we plan to run novel and state-of-the-art sampling and network mining algorithms such as matrix decompositions, computation of network statistics such as centralities and rankings, and the estimation of random graph model parameters on a large collection of over hundred large network datasets. The algorithms will be executed on the full large datasets, as well as on smaller samples of these. Thereby, it will be possible to identify or develop network mining algorithms which are stable to given sampling methods (i.e., provide the same results), as well as sampling algorithms which do not effect the outcome of a given data mining approach.

Last modified 7 years ago Last modified on May 3, 2012 1:27:42 PM