Graphlet decomposition of a weighted network
Methodology
2012-03-14 v1 Machine Learning
Social and Information Networks
Physics and Society
Abstract
We introduce the graphlet decomposition of a weighted network, which encodes a notion of social information based on social structure. We develop a scalable inference algorithm, which combines EM with Bron-Kerbosch in a novel fashion, for estimating the parameters of the model underlying graphlets using one network sample. We explore some theoretical properties of the graphlet decomposition, including computational complexity, redundancy and expected accuracy. We demonstrate graphlets on synthetic and real data. We analyze messaging patterns on Facebook and criminal associations in the 19th century.
Cite
@article{arxiv.1203.2821,
title = {Graphlet decomposition of a weighted network},
author = {Hossein Azari Soufiani and Edoardo M Airoldi},
journal= {arXiv preprint arXiv:1203.2821},
year = {2012}
}
Comments
25 pages, 4 figures, 3 tables