An empirical overview of nonlinearity and overfitting in machine learning using COVID-19 data

Vista sencilla de ítem
dc.creatorPeng, Yaohao
dc.creatorNagata, Mateus Hiro
dc.date.accessioned2020-07-27T19:54:10Z
dc.date.available2020-07-27T19:54:10Z
dc.date.created2020-10
dc.description.abstractenglishIn this paper, we applied support vector regression to predict the number of COVID-19 cases for the 12 most-affected countries, testing for different structures of nonlinearity using Kernel functions and analyzing the sensitivity of the models’ predictive performance to different hyperparameters settings using 3-D interpolated surfaces. In our experiment, the model that incorporates the highest degree of nonlinearity (Gaussian Kernel) had the best in-sample performance, but also yielded the worst out-of-sample predictions, a typical example of overfitting in a machine learning model. On the other hand, the linear Kernel function performed badly in-sample but generated the best out-of-sample forecasts. The findings of this paper provide an empirical assessment of fundamental concepts in data analysis and evidence the need for caution when applying machine learning models to support real-world decision making, notably with respect to the challenges arising from the COVID-19 pandemics.spa
dc.format.extent16 páginasspa
dc.format.mimetypeapplication/pdfspa
dc.identifier.doihttps://doi.org/10.1016/j.chaos.2020.110055spa
dc.identifier.issn0960-0779
dc.identifier.otherhttps://www.sciencedirect.com/science/article/pii/S0960077920304525?via%3Dihubspa
dc.identifier.urihttps://hdl.handle.net/20.500.12010/11217
dc.publisherChaos, Solitons and Fractalseng
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.sourcereponame:Expeditio Repositorio Institucional UJTLspa
dc.sourceinstname:Universidad de Bogotá Jorge Tadeo Lozanospa
dc.subjectPredicción series de tiempospa
dc.subject.keywordBias-variance dilemmaspa
dc.subject.keywordTime series predictionspa
dc.subject.keywordSupport vector machinespa
dc.subject.keywordStatistical learningspa
dc.subject.keywordHyperparameters and chaosspa
dc.subject.keywordEpidemic spreadingspa
dc.subject.lembSíndrome respiratorio agudo gravespa
dc.subject.lembCOVID-19spa
dc.subject.lembSARS-CoV-2spa
dc.subject.lembCoronavirusspa
dc.titleAn empirical overview of nonlinearity and overfitting in machine learning using COVID-19 dataspa
dc.type.hasversioninfo:eu-repo/semantics/acceptedVersionspa
dc.type.localArtículospa

Archivos

Bloque original

Mostrando 1 - 1 de 1
Miniatura
Nombre:
An-empirical-overview-of-nonlinearity-and-overfitting-i_2020_Chaos--Solitons.pdf
Tamaño:
3.78 MB
Formato:
Adobe Portable Document Format
Descripción:
Documento reservado
Descargar

Bloque de licencias

Mostrando 1 - 1 de 1
Miniatura
Nombre:
license.txt
Tamaño:
2.87 KB
Formato:
Item-specific license agreed upon to submission
Descripción:
Descargar

Colecciones

Documentos científicos relacionados a la COVID-19

Carrera 4 # 22-61 Teléfono: (+57) 601 242 7030 - 018000111022 Fax: (+57) 601 561 2107 Bogotá D.C., Colombia

Fundación Universitaria de Bogotá Jorge Tadeo Lozano | Vigilada Mineducación

Institución de educación superior privada, de utilidad común, sin ánimo de lucro y su carácter académico es el de Universidad

Reconocimiento personería jurídica: Resolución 2613 del 14 de agosto de 1959 Minjusticia.

Institución de Educación Superior sujeta a inspección y vigilancia por el Ministerio de Educación Nacional.

Software DSpace copyright © 2002-2026 LYRASIS