pone.0259203.s003.pdf (84.91 kB)
Relative importance of features in the classification process.
journal contribution
posted on 2021-11-04, 17:35 authored by Alberto Garcia-Zamalloa, Diego Vicente, Rafael Arnay, Arantzazu Arrospide, Jorge Taboada, Iván Castilla-Rodríguez, Urko Aguirre, Nekane Múgica, Ladislao Aldama, Borja Aguinagalde, Montserrat Jimenez, Edurne Bikuña, Miren Begoña Basauri, Marta Alonso, Emilio Perez-TralleroThe Random Forest classifier allows estimating the relative importance of each variable in the classification process. In view of the results, and in accordance with all the previously reported studies, in this classification process, “ADA” is the most important variable, followed by “age”. The rest of the variables are less useful, but we decided not to remove any of them a priori, but rather let each Machine Learning method assign a relative importance to each of them.
(PDF)
History
Usage metrics
Categories
Keywords
year prospective multitwo different classificationslow prevalence settinglow prevalence scenariocomposite reference standardroutine biochemical parametersparameters routinely measuredwhite cell countspleural fluid together98 %, whilst97 %, sensitivitypleural effusion includedmalignant may alsomachine learning algorithmspleural fluid combinedpleural fluidparameters included91 %,pleural tuberculosismalignant effusionlearning approachcell countalso allowedxlink ">validation predictionstest probabilitymarch 2013lymphocyte percentagelactate dehydrogenasejune 2020identify effusionsclinical practiceclassifier yieldedclass classifierbest resultbased classifiersadenosine deaminase80 %.
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC