Network Global Testing by Counting Graphlets

Jiashun Jin, Zheng Ke, Shengming Luo
; Proceedings of the 35th International Conference on Machine Learning, PMLR 80:2333-2341, 2018.

Abstract

Consider a large social network with possibly severe degree heterogeneity and mixed-memberships. We are interested in testing whether the network has only one community or there are more than one communities. The problem is known to be non-trivial, partially due to the presence of severe degree heterogeneity. We construct a class of test statistics using the numbers of short paths and short cycles, and the key to our approach is a general framework for canceling the effects of degree heterogeneity. The tests compare favorably with existing methods. We support our methods with careful analysis and numerical study with simulated data and a real data example.

Cite this Paper


BibTeX
@InProceedings{pmlr-v80-jin18b, title = {Network Global Testing by Counting Graphlets}, author = {Jin, Jiashun and Ke, Zheng and Luo, Shengming}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {2333--2341}, year = {2018}, editor = {Jennifer Dy and Andreas Krause}, volume = {80}, series = {Proceedings of Machine Learning Research}, address = {Stockholmsmässan, Stockholm Sweden}, month = {10--15 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v80/jin18b/jin18b.pdf}, url = {http://proceedings.mlr.press/v80/jin18b.html}, abstract = {Consider a large social network with possibly severe degree heterogeneity and mixed-memberships. We are interested in testing whether the network has only one community or there are more than one communities. The problem is known to be non-trivial, partially due to the presence of severe degree heterogeneity. We construct a class of test statistics using the numbers of short paths and short cycles, and the key to our approach is a general framework for canceling the effects of degree heterogeneity. The tests compare favorably with existing methods. We support our methods with careful analysis and numerical study with simulated data and a real data example.} }
Endnote
%0 Conference Paper %T Network Global Testing by Counting Graphlets %A Jiashun Jin %A Zheng Ke %A Shengming Luo %B Proceedings of the 35th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jennifer Dy %E Andreas Krause %F pmlr-v80-jin18b %I PMLR %J Proceedings of Machine Learning Research %P 2333--2341 %U http://proceedings.mlr.press %V 80 %W PMLR %X Consider a large social network with possibly severe degree heterogeneity and mixed-memberships. We are interested in testing whether the network has only one community or there are more than one communities. The problem is known to be non-trivial, partially due to the presence of severe degree heterogeneity. We construct a class of test statistics using the numbers of short paths and short cycles, and the key to our approach is a general framework for canceling the effects of degree heterogeneity. The tests compare favorably with existing methods. We support our methods with careful analysis and numerical study with simulated data and a real data example.
APA
Jin, J., Ke, Z. & Luo, S.. (2018). Network Global Testing by Counting Graphlets. Proceedings of the 35th International Conference on Machine Learning, in PMLR 80:2333-2341

Related Material