[R] Graph densification in large networks

[Lietz, Haiko]

I didn't remember Leskovec et al.'s work correctly. They don't compute the number of new nodes and new edges per unit time but just the number of nodes and edges. This makes it a lot easier.

But nevertheless, if I wanted to compute the number of new edges - what package would I use for large (and not necessarily sparse) graphs?

Haiko
 

-----Ursprüngliche Nachricht-----
Von: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] Im Auftrag von Lietz, Haiko
Gesendet: Montag, 10. März 2014 10:34
An: r-help at r-project.org
Betreff: [R] Graph densification in large networks

Hi all,

I have dynamic large 2-mode networks (1st mode ~50k nodes, 2nd mode ~500k nodes, timepoints ~100).

I want to identify the number of new nodes and new edges per timepoint in folded 1-mode networks.

In other words, I want to do measure graph densification as proposed by Leskovec et al. (http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.262.3092&rep=rep1&type=pdf)

What package should I use?

Best wishes

Haiko


Haiko Lietz
GESIS - Leibniz Institute for the Social Sciences
Department of Computational Social Science
Unter Sachsenhausen 6-8, D-50667 Köln
Tel: + 49 (0) 221 / 476 94 -223
eMail: haiko.lietz at gesis.org<mailto:haiko.lietz at gesis.org>
Web: http://www.gesis.org<http://www.gesis.org/>


	[[alternative HTML version deleted]]