|
Activity Number:
|
244
- Statistical methods for microbiome data analysis and beyond
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 11, 2021 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
Section on Statistics in Genomics and Genetics
|
|
Abstract #318567
|
|
|
Title:
|
Choice of Significance Level in a Community Detection Algorithm
|
|
Author(s):
|
Riddhi Pratim Ghosh* and Ian Barnett
|
|
Companies:
|
University of Pennsylvania and University of Pennsylvania
|
|
Keywords:
|
Community detection;
Sequential testing;
Significance;
Underfitting
|
|
Abstract:
|
Detecting network community structure is a common application in network science. While there have been numerous algorithms developed to estimate community structure, there is little available guidance and study of how to obtain the correct number of communities. Most algorithms rely on pre-specifying the number of communities or use an arbitrary stopping rule. We provide a less arbitrary means of determining the number of communities by selecting a nominal significance level for sequential community detection procedures that control the underfitting probability. We introduce an algorithm for specifying this significance level from a user-specified underfitting probability for sequential modularity maximization approach in a stochastic block model framework. We provide necessary conditions for the convergence of our algorithm. The performance of the proposed algorithm has been validated through extensive simulation study and a real data example.
|
Authors who are presenting talks have a * after their name.
