Cargando…

Comparison of Two Dimension-Reduction Methods for Network Simulation Models

Experimenters characterize the behavior of simulation models for data communications networks by measuring multiple responses under selected parameter combinations. The resulting multivariate data may include redundant responses reflecting aspects of a smaller number of underlying behaviors. Reducin...

Descripción completa

Detalles Bibliográficos
Autores principales: Mills, Kevin L., Filliben, James J.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: [Gaithersburg, MD] : U.S. Dept. of Commerce, National Institute of Standards and Technology 2011
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4551218/
https://www.ncbi.nlm.nih.gov/pubmed/26989599
http://dx.doi.org/10.6028/jres.116.020
_version_ 1782387544757895168
author Mills, Kevin L.
Filliben, James J.
author_facet Mills, Kevin L.
Filliben, James J.
author_sort Mills, Kevin L.
collection PubMed
description Experimenters characterize the behavior of simulation models for data communications networks by measuring multiple responses under selected parameter combinations. The resulting multivariate data may include redundant responses reflecting aspects of a smaller number of underlying behaviors. Reducing the dimension of multivariate responses can reveal the most significant model behaviors, allowing subsequent analyses to focus on one response per behavior. This paper investigates two methods for reducing dimension in multivariate data generated from simulation models. One method combines correlation analysis and clustering. The second method uses principal components analysis. We apply both methods to reduce a 22-dimensional dataset generated by a network simulator. We identify issues that an analyst must decide, and we compare the reductions suggested by the methods. We have used these methods to identify significant behaviors in simulated networks, and we suspect they may be applied to reduce the dimension of empirical data measured from real networks.
format Online
Article
Text
id pubmed-4551218
institution National Center for Biotechnology Information
language English
publishDate 2011
publisher [Gaithersburg, MD] : U.S. Dept. of Commerce, National Institute of Standards and Technology
record_format MEDLINE/PubMed
spelling pubmed-45512182016-03-17 Comparison of Two Dimension-Reduction Methods for Network Simulation Models Mills, Kevin L. Filliben, James J. J Res Natl Inst Stand Technol Article Experimenters characterize the behavior of simulation models for data communications networks by measuring multiple responses under selected parameter combinations. The resulting multivariate data may include redundant responses reflecting aspects of a smaller number of underlying behaviors. Reducing the dimension of multivariate responses can reveal the most significant model behaviors, allowing subsequent analyses to focus on one response per behavior. This paper investigates two methods for reducing dimension in multivariate data generated from simulation models. One method combines correlation analysis and clustering. The second method uses principal components analysis. We apply both methods to reduce a 22-dimensional dataset generated by a network simulator. We identify issues that an analyst must decide, and we compare the reductions suggested by the methods. We have used these methods to identify significant behaviors in simulated networks, and we suspect they may be applied to reduce the dimension of empirical data measured from real networks. [Gaithersburg, MD] : U.S. Dept. of Commerce, National Institute of Standards and Technology 2011 2011-10-01 /pmc/articles/PMC4551218/ /pubmed/26989599 http://dx.doi.org/10.6028/jres.116.020 Text en https://creativecommons.org/publicdomain/zero/1.0/ The Journal of Research of the National Institute of Standards and Technology is a publication of the U.S. Government. The papers are in the public domain and are not subject to copyright in the United States. Articles from J Res may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
spellingShingle Article
Mills, Kevin L.
Filliben, James J.
Comparison of Two Dimension-Reduction Methods for Network Simulation Models
title Comparison of Two Dimension-Reduction Methods for Network Simulation Models
title_full Comparison of Two Dimension-Reduction Methods for Network Simulation Models
title_fullStr Comparison of Two Dimension-Reduction Methods for Network Simulation Models
title_full_unstemmed Comparison of Two Dimension-Reduction Methods for Network Simulation Models
title_short Comparison of Two Dimension-Reduction Methods for Network Simulation Models
title_sort comparison of two dimension-reduction methods for network simulation models
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4551218/
https://www.ncbi.nlm.nih.gov/pubmed/26989599
http://dx.doi.org/10.6028/jres.116.020
work_keys_str_mv AT millskevinl comparisonoftwodimensionreductionmethodsfornetworksimulationmodels
AT fillibenjamesj comparisonoftwodimensionreductionmethodsfornetworksimulationmodels