Graduate Thesis Or Dissertation
 

Toward a flexible and accurate approach to null network sampling

Public Deposited

Downloadable Content

Download PDF
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/05742105j

Descriptions

Attribute NameValues
Creator
Abstract
  • Null networks are a type of random graph that is favored for the analysis of a wide variety of real-world networks, including gene-regulatory networks, food webs, and species co-occurrence matrices. As a hypothesis-generating tool, null networks are invaluable because they can reveal network motifs and unusual large-scale properties of networks that are associated with the degree sequence in expectation and would go undetected without a background model. In order to provide useful insight, a null network configuration model must be sampled uniformly and quickly enough to generate a reasonably large sample of random networks. In this work, we extend existing knowledge on the performance of two well-known edge-swapping Markov Chain algorithms for random graph sampling with experimental evidence using artificial graphs a variety of degree distributions, and demonstrate some situations where an optimized stub-matching algorithm can complement their performance on difficult degree distributions. We additionally demonstrate how two criteria can be helpful in determining when sampling uniformity is inadequate.
License
Resource Type
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Academic Affiliation
Rights Statement
Publisher
Peer Reviewed
Language

Relationships

Parents:

This work has no parents.

In Collection:

Items

Thumbnail Title Date Uploaded Visibility Actions
file details ZacharyBright2023.pdf 2023-04-03 Public Download