Feature selection in the reconstruction of complex network representations of spectral data.
Complex networks have been extensively used in the last decade to characterize and analyze complex systems, and they have been recently proposed as a novel instrument for the analysis of spectra extracted from biological samples. Yet, the high number of measurements composing spectra, and the conseq...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Public Library of Science (PLoS)
2013-01-01
|
Series: | PLoS ONE |
Online Access: | http://europepmc.org/articles/PMC3753346?pdf=render |
id |
doaj-becb9dbe76ce4a768fc7e704b827bd2c |
---|---|
record_format |
Article |
spelling |
doaj-becb9dbe76ce4a768fc7e704b827bd2c2020-11-24T22:00:28ZengPublic Library of Science (PLoS)PLoS ONE1932-62032013-01-0188e7204510.1371/journal.pone.0072045Feature selection in the reconstruction of complex network representations of spectral data.Massimiliano ZaninErnestina MenasalvasStefano BoccalettiPedro SousaComplex networks have been extensively used in the last decade to characterize and analyze complex systems, and they have been recently proposed as a novel instrument for the analysis of spectra extracted from biological samples. Yet, the high number of measurements composing spectra, and the consequent high computational cost, make a direct network analysis unfeasible. We here present a comparative analysis of three customary feature selection algorithms, including the binning of spectral data and the use of information theory metrics. Such algorithms are compared by assessing the score obtained in a classification task, where healthy subjects and people suffering from different types of cancers should be discriminated. Results indicate that a feature selection strategy based on Mutual Information outperforms the more classical data binning, while allowing a reduction of the dimensionality of the data set in two orders of magnitude.http://europepmc.org/articles/PMC3753346?pdf=render |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Massimiliano Zanin Ernestina Menasalvas Stefano Boccaletti Pedro Sousa |
spellingShingle |
Massimiliano Zanin Ernestina Menasalvas Stefano Boccaletti Pedro Sousa Feature selection in the reconstruction of complex network representations of spectral data. PLoS ONE |
author_facet |
Massimiliano Zanin Ernestina Menasalvas Stefano Boccaletti Pedro Sousa |
author_sort |
Massimiliano Zanin |
title |
Feature selection in the reconstruction of complex network representations of spectral data. |
title_short |
Feature selection in the reconstruction of complex network representations of spectral data. |
title_full |
Feature selection in the reconstruction of complex network representations of spectral data. |
title_fullStr |
Feature selection in the reconstruction of complex network representations of spectral data. |
title_full_unstemmed |
Feature selection in the reconstruction of complex network representations of spectral data. |
title_sort |
feature selection in the reconstruction of complex network representations of spectral data. |
publisher |
Public Library of Science (PLoS) |
series |
PLoS ONE |
issn |
1932-6203 |
publishDate |
2013-01-01 |
description |
Complex networks have been extensively used in the last decade to characterize and analyze complex systems, and they have been recently proposed as a novel instrument for the analysis of spectra extracted from biological samples. Yet, the high number of measurements composing spectra, and the consequent high computational cost, make a direct network analysis unfeasible. We here present a comparative analysis of three customary feature selection algorithms, including the binning of spectral data and the use of information theory metrics. Such algorithms are compared by assessing the score obtained in a classification task, where healthy subjects and people suffering from different types of cancers should be discriminated. Results indicate that a feature selection strategy based on Mutual Information outperforms the more classical data binning, while allowing a reduction of the dimensionality of the data set in two orders of magnitude. |
url |
http://europepmc.org/articles/PMC3753346?pdf=render |
work_keys_str_mv |
AT massimilianozanin featureselectioninthereconstructionofcomplexnetworkrepresentationsofspectraldata AT ernestinamenasalvas featureselectioninthereconstructionofcomplexnetworkrepresentationsofspectraldata AT stefanoboccaletti featureselectioninthereconstructionofcomplexnetworkrepresentationsofspectraldata AT pedrosousa featureselectioninthereconstructionofcomplexnetworkrepresentationsofspectraldata |
_version_ |
1725844375357882368 |